ECONOMICS 351* -- Stata 10 Tutorial 7. Stata 10 Tutorial 7. TOPIC: Estimation and Hypothesis Testing in Multiple Linear Regression Models

Size: px
Start display at page:

Download "ECONOMICS 351* -- Stata 10 Tutorial 7. Stata 10 Tutorial 7. TOPIC: Estimation and Hypothesis Testing in Multiple Linear Regression Models"

Transcription

1 ECONOMICS 5* -- Stata 0 Tutoral 7 Stata 0 Tutoral 7 TOPIC: Estmaton and Hypothess Testng n Multple Lnear Regresson Models DATA: autodta (a Stata-format data fle you created n Stata Tutoral ) TASKS: Stata Tutoral 7 ntroduces you to OLS estmaton of multple lnear regresson models contanng two or more regressors, and demonstrates how to perform varous common types of hypothess tests n such models It s ntended to gve you some practce n formulatng multple lnear regresson models and n computng F-tests and t-tests of lnear coeffcent restrctons on such models It ntroduces you to the specfcaton and nterpretaton of multple lnear regresson models that nclude as regressors polynomal terms n explanatory varables and nteracton terms between contnuous explanatory varables It also ntroduces the Stata lncom command, a post-estmaton command that computes lnear combnatons of coeffcent estmates, specfcally pont estmates, standard errors, t- statstcs, two-tal p-values, and two-sded confdence ntervals for specfed lnear combnatons of coeffcent estmates The Stata commands that consttute the prmary subject of ths tutoral are: regress test lncom Used to perform OLS estmaton of multple lnear regresson models Used to compute F-tests of lnear coeffcent equalty restrctons after OLS estmaton Used after estmaton to compute lnear combnatons of coeffcent estmates and assocated statstcs The Stata statstcal functons ntroduced n ths tutoral are: ttal(df, t 0 ) nvttal(df, p) Ftal(df, df, F 0 ) nvftal(df, df, α) Computes p-values of calculated t-statstcs Computes rght-tal crtcal values of t-dstrbutons Computes p-values of calculated F-statstcs Computes rght-tal crtcal values of F-dstrbutons NOTE: Stata commands are case senstve All Stata command names must be typed n the Command wndow n lower case letters ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

2 ECONOMICS 5* -- Stata 0 Tutoral 7 Overvew of Regresson Models Used n ths Tutoral The followng multple lnear regresson models are estmated and tested n ths Stata tutoral prce β + β weght + β mpg + u () 0 0 ln( prce ) α + α ln(weght ) + α ln(mpg ) + u () prce β + β weght + β mpg + β weght + u (7) 0 0 prce β + β weght + β mpg + β weght + β mpg + u () prce β0 + βweght + βmpg + βweght + β4mpg + β5weghtmpg + u () 4 Preparng for Your Stata Sesson Before begnnng your Stata sesson, use Wndows Explorer to copy the Stataformat dataset autodta to the Stata workng drectory on the C:-drve or D:-drve of the computer at whch you are workng On the computers n Dunnng 50, the default Stata workng drectory s usually C:\data On the computers n MC B, the default Stata workng drectory s usually D:\courses Start Your Stata Sesson To start your Stata sesson, double-clck on the Stata 0 con n the Wndows desktop After you double-clck the Stata 0 con, you wll see the now famlar screen of four Stata wndows ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

3 ECONOMICS 5* -- Stata 0 Tutoral 7 Record Your Stata Sesson and Stata Commands -- log usng, cmdlog usng To record your Stata sesson, ncludng all the Stata commands you enter and the results (output) produced by these commands, make a log fle named 5tutoral7log To open (begn) the log fle 5tutoral7log, enter n the Command wndow: log usng 5tutoral7log Ths command opens a fle called 5tutoral7log n the current Stata workng drectory Remember that once you have opened the 5tutoral7log fle, a copy of all the commands you enter durng your Stata sesson and of all the results they produce s recorded n that 5tutoral7log fle An alternatve way to open the log fle 5tutoral7log s to clck on the Log button; clck on Save as type: and select Log (*log); clck on the Fle name: box and type the fle name 5tutoral7; and clck on the Save button To record only the Stata commands you type durng your Stata sesson, use the Stata cmdlog usng command To start (open) the command log fle 5tutoral7txt, enter n the Command wndow: cmdlog usng 5tutoral7 Ths command opens a plan text-format (ASCII) fle called 5tutoral7txt n the current Stata workng drectory All commands you enter durng your Stata sesson are recorded n ths fle Loadng a Stata-Format Dataset nto Stata -- use Load, or read, nto memory the dataset you are usng To load the Stata-format data fle autodta nto memory, enter n the Command wndow: use auto Ths command loads nto memory the Stata-format dataset autodta ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

4 ECONOMICS 5* -- Stata 0 Tutoral 7 Famlarze Yourself wth the Current Dataset To famlarze (or re-famlarze) yourself wth the contents of the current dataset, type n the Command wndow the followng commands: descrbe, short descrbe summarze Computng lnear combnatons of coeffcent estmates -- lncom Basc Syntax: lncom exp [, level(#)] where exp s a user-specfed lnear combnaton of coeffcent estmates The lncom command computes pont estmates, standard errors, t-statstcs, p- values, and two-sded confdence ntervals for a specfed lnear combnaton of coeffcent estmates from the most recently executed model estmaton command, such as the regress command Note that the lnear combnaton specfed n exp cannot contan an equalty sgn () The level(#) opton on the lncom command specfes the confdence level to be used n computng the two-sded confdence nterval for the specfed lnear combnaton of regresson coeffcents The value of # specfes the desred confdence level n percentage ponts; the default confdence level s 95 percent (e, α 095) Examples: The lncom command s most useful n multple regresson models contanng two or more explanatory varables Several examples of ts use are gven n subsequent sectons of ths tutoral Estmaton and Inference n Multple Lnear Regresson Models: Introducton Ths secton ntroduces you to OLS estmaton of multple lnear regresson models contanng two or more regressors It also demonstrates how to perform varous common types of hypothess tests n such models ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 4 of 9 pages

5 ECONOMICS 5* -- Stata 0 Tutoral 7 Consder the followng multple lnear regresson model for prce : prce β + β weght + β mpg + u () 0 where prce the prce of the -th car (n US dollars); weght the weght of the -th car (n pounds); and mpg the mles per gallon for the -th car (n mles per gallon) Note: For PRE (), the number of regresson coeffcents s K Regresson equaton () contans two explanatory varables, weght and mpg The margnal effect of weght n equaton () s: prce weght β a constant The margnal effect of mpg n equaton () s: prce mpg β a constant To estmate model () by OLS for the full sample of N 74 sample observatons n dataset autodta, enter n the Command wndow: regress prce weght mpg Testng the ndvdual sgnfcance of the slope coeffcent estmates Indvdual sgnfcance tests are two-tal tests of the followng statstcal hypothess: H 0 : β j 0 versus H : β j 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 5 of 9 pages

6 ECONOMICS 5* -- Stata 0 Tutoral 7 Such ndvdual sgnfcance tests can be performed usng ether a two-tal t-test or an F-test The feasble test statstcs for such ndvdual sgnfcance tests are ether the t- statstc for ˆβ j gven by t(ˆ β j ) βˆ j β j sê(ˆ β ) j ~ t[n K] or the F-statstc for ˆβ j gven by ( βˆ β ) j j F(ˆ β j) ~ F[, N K] Vâr(ˆ β ) j Example : In regresson equaton (), perform a two-tal test of the hypothess H 0 : β 0 versus H : β 0 The null hypothess H 0 : β 0 mples that the margnal effect of the explanatory varable weght equals zero n PRE (): H 0 : β 0 prce β weght 0 Ths hypothess can be tested usng ether a two-tal t-test or an equvalent F-test A two-tal t-test: Under the null hypothess H 0 : β 0, the sample value of the t-statstc for β ˆ s t 0 βˆ ˆ ˆ β β 0 β (ˆ β ) sê(ˆ β ) sê(ˆ β ) sê(ˆ β ) ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 6 of 9 pages

7 ECONOMICS 5* -- Stata 0 Tutoral 7 The null dstrbuton of t β 0 (ˆ ) s: t 0 (ˆ β ) ~ t[n K] t[74 ] t[7] Compute the sample value of the t-statstc Enter the commands: scalar t0b _b[weght]/_se[weght] scalar lst t0b Compute the two-tal p-value of the calculated t-statstc Enter the command: dsplay *ttal(7, abs(t0b)) Another way to do what the precedng three commands have done s to use the Stata lncom command to compute both the sample value of the t-statstc and ts two-tal p-value Enter the command: lncom _b[weght] What nference would you draw from the outcome of ths two-tal t-test? Compute the two-tal 00α percent crtcal values t [7] α / and t [7 α / ] of the t[7]-dstrbuton Use the nvttal(df, p) functon, where p α/ At the % sgnfcance level for whch α 00, α/ 00/ 0005 Enter the commands: dsplay nvttal(7, 0005) (returns value of t [7] ) dsplay -nvttal(7, 0005) (returns value of t 005 [7] ) At the 5% sgnfcance level for whch α 005, α/ 005/ 005 Enter the commands: dsplay nvttal(7, 005) (returns value of t 0 05 [7] ) dsplay -nvttal(7, 005) (returns value of t 05 [7] ) 0 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 7 of 9 pages

8 ECONOMICS 5* -- Stata 0 Tutoral 7 At the 0% sgnfcance level for whch α 00, α/ 00/ 005 Enter the commands: dsplay nvttal(7, 005) (returns value of t 0 05 [7] ) dsplay -nvttal(7, 005) (returns value of t 05 [7] ) 0 An F-test: Under the null hypothess H 0 : β 0, the sample value of the F- statstc for ˆβ s ( βˆ β ) ( βˆ 0) F 0(βˆ ) βˆ βˆ Vâr(βˆ ) Vâr(βˆ ) Vâr(βˆ ) sê(βˆ ) The null dstrbuton of F β 0 (ˆ ) s: F0 (ˆ β ) ~ F[, N K] F[, 74 ] F[, 7] Compute the sample value of the F-statstc Enter the commands: scalar F0b _b[weght]^/_se[weght]^ scalar lst F0b Compute the p-value of the calculated F-statstc Enter the command: dsplay Ftal(, 7, F0b) Another way to do what the precedng three commands have done s to use the Stata test command to compute both the sample value of the F-statstc and ts p-value Enter ether one of the followng test commands: test weght 0 test weght What nference would you draw from the outcome of ths F-test? Compute the 00α percent crtcal value F α [, 7] of the F[, 7]-dstrbuton ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 8 of 9 pages

9 ECONOMICS 5* -- Stata 0 Tutoral 7 Use the nvftal(df, df, α) statstcal functon to compute crtcal values of the F[df, df ]-dstrbuton, where df s the numerator degrees of freedom, df s the denomnator degrees of freedom, and α s the chosen sgnfcance level In the present case, df (the number of coeffcent restrctons specfed by the null hypothess H 0 ) and df 7 (the degrees of freedom for the RSS of the unrestrcted regresson equaton ()) The followng commands compute and dsplay the crtcal values of the F[,7] dstrbuton for sgnfcance levels α 00, 005, 00 At the % sgnfcance level, α 00 Enter the command: dsplay nvftal(,7, 00) (returns value of F 0 [, 7] ) At the 5% sgnfcance level, α 005 Enter the command: dsplay nvftal(,7, 005) (returns value of F 05 [, 7] ) At the 0% sgnfcance level, α 00 Enter the command: dsplay nvftal(,7, 00) (returns value of F 0 [, 7] ) These are the crtcal values of F α [, 7] for testng H 0 : β 0 aganst H : β 0 Would you reject the null hypothess β 0 n favor of the alternatve hypothess β 0 at the 0 percent sgnfcance level, at the 5 percent sgnfcance level, and/or at the percent sgnfcance level? Do the two tests the two-tal t-test and the F-test yeld the same nference? Example : In regresson equaton (), perform a two-tal test of the hypothess H 0 : β 0 versus H : β 0 The null hypothess H 0 : β 0 mples that the margnal effect of the explanatory varable mpg equals zero n PRE (): H 0 : β 0 prce β mpg 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 9 of 9 pages

10 ECONOMICS 5* -- Stata 0 Tutoral 7 Ths hypothess can be tested usng ether a two-tal t-test or an equvalent F-test A two-tal t-test: Under the null hypothess H 0 : β 0, the sample value of the t-statstc for β ˆ s t 0 βˆ ˆ ˆ β β 0 β (ˆ β ) sê(ˆ β ) sê(ˆ β ) sê(ˆ β ) The null dstrbuton of t β 0 (ˆ ) s: t 0 (ˆ β ) ~ t[n K] t[74 ] t[7] Compute the sample value of the t-statstc Enter the commands: scalar t0b _b[mpg]/_se[mpg] scalar lst t0b Compute the two-tal p-value of the calculated t-statstc Enter the command: dsplay *ttal(7, abs(t0b)) Another way to do what the precedng three commands have done s to use the Stata lncom command to compute both the sample value of the t-statstc and ts two-tal p-value Enter the command: lncom _b[mpg] What nference would you draw from the outcome of ths two-tal t-test? Re-compute the two-tal 00α percent crtcal values t [7] α / and t [7 α / ] of the t[7]-dstrbuton at sgnfcance levels α 00, 005, 00 Use the nvttal(df, p) functon wth p α/ Enter the commands: dsplay nvttal(7, 0005) (returns value of t [7] ) dsplay -nvttal(7, 0005) (returns value of t 005 [7] ) 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 0 of 9 pages

11 ECONOMICS 5* -- Stata 0 Tutoral 7 dsplay nvttal(7, 005) (returns value of t 0 05 [7] ) dsplay -nvttal(7, 005) (returns value of t 05 [7] ) dsplay nvttal(7, 005) (returns value of t 0 05 [7] ) dsplay -nvttal(7, 005) (returns value of t 05 [7] ) 0 0 An F-test: Under the null hypothess H 0 : β 0, the sample value of the F- statstc for ˆβ s ( βˆ β ) ( βˆ 0) F 0(βˆ ) βˆ βˆ Vâr(βˆ ) Vâr(βˆ ) Vâr(βˆ ) sê(βˆ ) The null dstrbuton of F β 0 (ˆ ) s: F0 (ˆ β ) ~ F[, N K] F[, 74 ] F[, 7] Compute the sample value of the F-statstc Enter the commands: scalar F0b _b[mpg]^/_se[mpg]^ scalar lst F0b Compute the p-value of the calculated F-statstc Enter the command: dsplay Ftal(, 7, F0b) Another way to do what the precedng three commands have done s to use the Stata test command to compute both the sample value of the F-statstc and ts p-value Enter ether one of the followng test commands: test mpg 0 test mpg What nference would you draw from the outcome of ths F-test? ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

12 ECONOMICS 5* -- Stata 0 Tutoral 7 Re-compute the 00α percent crtcal values F α [, 7] of the F[, 7]- dstrbuton at sgnfcance levels α 00, 005, 00 At the % sgnfcance level, α 00 Enter the command: dsplay nvftal(,7, 00) (returns value of F 0 [, 7] ) At the 5% sgnfcance level, α 005 Enter the command: dsplay nvftal(,7, 005) (returns value of F 05 [, 7] ) At the 0% sgnfcance level, α 00 Enter the command: dsplay nvftal(,7, 00) (returns value of F 0 [, 7] ) These are the crtcal values of F α [, 7] for testng H 0 : β 0 aganst H : β 0 Would you reject the null hypothess β 0 n favor of the alternatve hypothess β 0 at the 0 percent sgnfcance level, at the 5 percent sgnfcance level, and/or at the percent sgnfcance level? Do the two tests the two-tal t-test and the F-test yeld the same nference? Testng the jont sgnfcance of all slope coeffcent estmates A jont sgnfcance test of all slope coeffcent estmates s a test the hypothess that the two slope coeffcents n PRE () are jontly equal to zero The null and alternatve hypotheses are: H 0 : β 0 j, β 0 and β 0 () j H : β 0 j, β 0 and/or β 0 j The followng sequence of test commands wth the accumulate opton wll perform a jont F-test of the null hypothess H 0 n () Enter the test commands: test weght 0 test mpg 0, accum ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

13 ECONOMICS 5* -- Stata 0 Tutoral 7 Note that the accumulate opton can be abbrevated as accum In fact, t can even be abbrevated as a Ths seres of test commands computes the sample value of the F-statstc and ts p-value Alternatvely, a sngle test command can be used to compute ths jont F-test when the null hypothess specfes zero or excluson restrctons on two or more regresson coeffcents, as t does n the case of the null and alternatve hypotheses n () above Enter the command: test weght mpg Ths test command performs a jont F-test of the null hypothess that the regresson coeffcent of the regressor weght equals zero and the regresson coeffcent of the regressor mpg equals zero Use the nvftal(df, df, α) statstcal functon to compute crtcal values of the F[df, df ]-dstrbuton, where df s the numerator degrees of freedom, df s the denomnator degrees of freedom, and α s the chosen sgnfcance level In the present case, df (the number of coeffcent restrctons specfed by the null hypothess H 0 ) and df 7 (the degrees of freedom for the RSS of the unrestrcted regresson equaton ()) The followng commands compute and dsplay the crtcal values of the F[,7] dstrbuton for sgnfcance levels α 00, 005, 00 At the % sgnfcance level, α 00 Enter the command: dsplay nvftal(,7, 00) (returns value of F 0 [, 7] ) At the 5% sgnfcance level, α 005 Enter the command: dsplay nvftal(,7, 005) (returns value of F 05 [, 7] ) At the 0% sgnfcance level, α 00 Enter the command: dsplay nvftal(,7, 00) (returns value of F 0 [, 7] ) These are the crtcal values approprate for testng hypothess () above ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

14 ECONOMICS 5* -- Stata 0 Tutoral 7 Would you reject the null hypothess n () at the 0 percent sgnfcance level, at the 5 percent sgnfcance level, and/or at the percent sgnfcance level? Testng Equalty of Slope Coeffcents n Multple Lnear Regresson Models: Introducton to the lncom Command Recall the multple lnear regresson model for prce consdered n the prevous secton: prce β + β weght + β mpg + u () 0 In ths regresson equaton, there s n general no reason to thnk that the slope coeffcents β and β wll be equal, because the explanatory varables weght and mpg wll usually have dfferent unts of measurement However, consder the followng log-log regresson model for car prces: ln( prce ) α + α ln(weght ) + α ln(mpg ) + u () 0 In the log-log regresson model (), the slope coeffcents α and α are partal elastcty coeffcents, and hence do not depend on the unts of measurement for prce, weght and mpg In fact, ln(prce ) α the partal elastcty of prce wrt weght ; ln(weght ) ln(prce ) α the partal elastcty of prce wrt mpg ln(mpg ) In log-log models such as that gven by populaton regresson equaton (), testng the equalty of pars of slope coeffcents makes perfect sense, snce these slope coeffcents are comparable n terms of ther magntude Before estmatng regresson equaton (), t s necessary to create the varables ln( prce ), ln( weght ) and ln( mpg ) Enter the commands: generate lnprce ln(prce) generate lnweght ln(weght) generate lnmpg ln(mpg) ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 4 of 9 pages

15 ECONOMICS 5* -- Stata 0 Tutoral 7 Estmate by OLS regresson equaton () Enter the command: regress lnprce lnweght lnmpg Test : Use the results of OLS estmaton of the log-log model () to test the proposton that the partal elastcty of prce wth respect to weght s equal to the partal elastcty of prce wth respect to mpg The null and alternatve hypotheses are: H 0 : H : α α 0 (4) α α α α 0 α α Use a Stata lncom command to compute a two-tal t-test of the null hypothess (4) above Enter the command: lncom _b[lnweght] - _b[lnmpg] Carefully nspect the results of ths lncom command State the nference you would draw from ths t-test Now use a Stata test command to compute an F-test of the null hypothess (4) Enter ether of the followng commands: test lnweght - lnmpg 0 test lnweght lnmpg Carefully nspect the results of ths test command How do you know that ths F-test yelds the same nference as the two-tal t-test of the same null hypothess (4)? Test : Use the results of OLS estmaton of the log-log model () to test the proposton that the partal elastcty of prce wth respect to weght s equal n magntude but opposte n sgn to the partal elastcty of prce wth respect to mpg ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 5 of 9 pages

16 ECONOMICS 5* -- Stata 0 Tutoral 7 The null and alternatve hypotheses for ths test are: H 0 : H : α α + α 0 (5) α α + α 0 α α Use a Stata lncom command to compute a two-tal t-test of the null hypothess above Enter the command: lncom _b[lnweght] + _b[lnmpg] Carefully nspect the results of ths lncom command State the nference you would draw from ths t-test Use a Stata test command to compute an F-test of the null hypothess (5) Enter ether of the followng commands: test lnweght + lnmpg 0 test lnweght -lnmpg Carefully nspect the results of ths test command How do you know that ths F-test yelds the same nference as the two-tal t-test of the same null hypothess (5)? Interacton Varables: Polynomal Terms n Explanatory Varables Nature: Polynomal terms refer to nteger powers of explanatory varables For example, the second-order polynomal term n the varable X j s X j -squared, ; the thrd-order polynomal term n the varable X j s X j -cubed, X j Usage: Polynomal terms n explanatory varables are one way of allowng for nonlnear relatonshps between the dependent varable Y and ndvdual explanatory varables such as X j Consder the followng regresson equaton: X j Y β + β X + β X + β X + u (6) 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 6 of 9 pages

17 ECONOMICS 5* -- Stata 0 Tutoral 7 Compare the margnal effects on Y of the two explanatory varables X and X n equaton (6) The margnal effect of X n equaton (6) s: Y X β + β X a lnear functon of X The margnal effect of X n equaton (6) s: Y X β a constant Example : a polynomal n one explanatory varable, weght Consder the followng multple regresson equaton: prce β + β weght + β mpg + β weght + u (7) 0 where prce the prce of the -th car (n US dollars); weght the weght of the -th car (n pounds); weght the square of weght ; and mpg the mles per gallon for the -th car (n mles per gallon) Note: For PRE (7), the number of regresson coeffcents s K 4 Regresson equaton (7) contans two explanatory varables, weght and mpg, but three regressors weght, weght, and mpg Margnal Effects of the Explanatory Varables The margnal effect of weght n equaton (7) s: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 7 of 9 pages

18 ECONOMICS 5* -- Stata 0 Tutoral 7 prce weght β + β weght a lnear functon of weght The margnal effect of mpg n equaton (7) s: prce mpg β a constant Before estmatng regresson equaton (7), t s necessary to create the varable weght Enter the command: generate weghtsq weght^ Now estmate by OLS regresson equaton (7) Enter the command: regress prce weght mpg weghtsq Testng the margnal effects of ndvdual explanatory varables Test : Test the hypothess that the margnal effect of weght on prce s zero for all cars n regresson equaton (7) The margnal effect of weght on prce s obtaned by partally dfferentatng regresson equaton (7) wth respect to weght : prce weght β + β weght (,, N) Suffcent condtons for prce weght 0 for all are β 0 and β 0 We therefore want to test these two coeffcent excluson restrctons The null and alternatve hypotheses are: H 0 : β 0 j, β 0 and β 0 (8) j H : β 0 j, β 0 and/or β 0 j ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 8 of 9 pages

19 ECONOMICS 5* -- Stata 0 Tutoral 7 The followng test command computes an F-test of H 0 aganst H Enter the command: test weght weghtsq Inspect the results generated by ths command State the nference you would draw from ths test Note: The above test command s equvalent to the followng sequence of test commands wth the accumulate opton: test weght 0 test weghtsq 0, accum Use the followng commands to compute the crtcal values of the F[,70] dstrbuton for sgnfcance levels α 00, 005, 00: dsplay nvftal(,70, 00) (returns value of F 0 0 [, 70] ) dsplay nvftal(,70, 005) (returns value of F 0 05 [, 70] ) dsplay nvftal(,70, 00) (returns value of F 0 [, 70] ) These are the crtcal values of the F[, 70]-dstrbuton approprate for testng hypothess (8) above Would you reject the null hypothess n (8) at the 0 percent sgnfcance level, at the 5 percent sgnfcance level, and/or at the percent sgnfcance level? Test 4: Test the hypothess that the margnal effect of weght on prce s constant n regresson equaton (7) e, that t does not depend on weght The necessary and suffcent condton for prce weght n equaton (7) to be constant s β 0 We therefore want to test ths coeffcent restrcton The null and alternatve hypotheses are: H 0 : β 0 (9) H : β 0 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 9 of 9 pages

20 ECONOMICS 5* -- Stata 0 Tutoral 7 One way to test ths hypothess s to use the t-rato for ˆβ that s prnted by the regress command used to estmate equaton (7) The t-rato for ˆβ s gven n general by the expresson t βˆ (ˆ β ) ~ t[n K] t[74 4] t[70] under H 0 : β 0 sê(ˆ β ) 0 Alternatvely, you can calculate the above t-rato for by enterng the followng lncom command: lncom _b[weghtsq] ˆβ and ts two-tal p-value From the results of OLS estmaton of equaton (7), state the nference you would draw from ths t-test A second way to test ths hypothess s to compute an F-test of H 0 : β 0 aganst H : β 0 The sample value of the F-statstc for ths test s gven by the expresson ˆ β F0 (ˆ β ) ~ F[, N K] F[, 74 4] F[, 70] under H 0 : β 0 Vâr(ˆ β ) Enter ether of the followng test commands: test weghtsq test weghtsq 0 State the nference you would draw from ths F-test Compare the results of ths F-test wth those of the t-test performed above Are these two tests equvalent? Why or why not? Test 5: Test the hypothess that the margnal effect of mpg on prce s zero n regresson equaton (7), e, that prce does not depend on mpg The margnal effect of mpg on prce s obtaned by partally dfferentatng regresson equaton (7) wth respect to mpg : ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 0 of 9 pages

21 ECONOMICS 5* -- Stata 0 Tutoral 7 prce mpg β a constant (,, N) The necessary and suffcent condton for prce mpg n equaton (7) to equal zero s β 0 We therefore want to test ths coeffcent restrcton The null and alternatve hypotheses are: H 0 : β 0 (0) H : β 0 One way to test hypothess (0) s to use the t-rato for ˆβ that s prnted by the regress command used to estmate equaton (7) The t-rato for ˆβ s gven n general by the expresson t βˆ (ˆ β ) ~ t[n K] t[74 4] t[70] under H 0 : β 0 sê(ˆ β ) 0 Alternatvely, you can calculate the above t-rato for command: lncom _b[mpg] ˆβ by enterng the lncom From the results of OLS estmaton of equaton (7), state the nference you would draw from ths t-test A second way to test hypothess (0) s to perform an F-test of the sngle coeffcent restrcton specfed by the null hypothess H 0 : β 0 n (0) Enter ether of the followng test commands: test mpg test mpg 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

22 ECONOMICS 5* -- Stata 0 Tutoral 7 Note: The sample value of the F-statstc computed by the above test command equals the square of the sample value of the t-statstc for the null hypothess H 0 n (0): ( βˆ b ) ( βˆ 0) ( βˆ ) βˆ [ ] 0 ) t 0 ) F (βˆ Vâr(βˆ ) Vâr(βˆ ) Vâr(βˆ ) sê(βˆ ) Use the followng commands to verfy these relatonshps between t (ˆ β 0 ) and (ˆ ): F β 0 scalar t0b _b[mpg]/_se[mpg] scalar varb _se[mpg]^ scalar F0b _b[mpg]^/varb dsplay t0b dsplay sqrt(f0b) dsplay t0b^ dsplay F0b Use the followng commands to verfy that the two-tal p-value of the calculated t-statstc t (ˆ β 0 ) equals the p-value of the calculated F-statstc (ˆ ): F β 0 dsplay *ttal(70, abs(t0b)) dsplay Ftal(, 70, F0b) Fnally, calculate and compare the two-tal % crtcal value of the t[70]- dstrbuton and the % crtcal value of the F[, 70]-dstrbuton Enter the dsplay commands: dsplay nvttal(70, 0005)^ dsplay nvftal(, 70, 00) The results of these commands can be used to verfy that the α-level crtcal value of the F[, 70]-dstrbuton equals the square of the α/ crtcal value of the t[70] dstrbuton, e, that 0 0[, 70] t 0005[70] ( F ) or n general that F [, N K] t [N K α (βˆ ( ) α / ] ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

23 ECONOMICS 5* -- Stata 0 Tutoral 7 Example : polynomals n two explanatory varables, weght and mpg Consder the followng multple regresson equaton: prce β + β weght + β mpg + β weght + β mpg + u () 0 4 where mpg the square of mpg Note: For PRE (), the number of regresson coeffcents s K 5 Regresson equaton () contans two explanatory varables, weght and mpg, but four regressors weght, weght, mpg, and mpg Margnal Effects of the Explanatory Varables The margnal effect of weght n equaton () s: prce weght β + β weght a lnear functon of weght The margnal effect of mpg n equaton () s: prce mpg β + β 4 mpg a lnear functon of mpg Before estmatng regresson equaton (), t s necessary to create the varable Enter the command: mpg gen mpgsq mpg^ Now estmate by OLS regresson equaton () Enter the command: regress prce weght mpg weghtsq mpgsq ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

24 ECONOMICS 5* -- Stata 0 Tutoral 7 Test 6: Test the hypothess that the margnal effect of weght on prce s zero n regresson equaton () The margnal effect of weght on prce s obtaned by partally dfferentatng regresson equaton () wth respect to weght : prce weght β + β weght Suffcent condtons for prce weght 0 for all are β 0 and β 0 We therefore want to test these two coeffcent excluson restrctons on regresson equaton () The null and alternatve hypotheses are: H 0 : β 0 and β 0 H : β 0 and/or β 0 The followng test command computes an F-test of H 0 aganst H Enter the command: test weght weghtsq Inspect the results generated by ths command State the nference you would draw from ths test Test 7: Use a lncom command to compute the estmated value of the margnal effect of weght n equaton () for a car that weghs,000 pounds, and to perform a two-tal t-test of the null hypothess that the margnal effect of weght n equaton () for a car that weghs,000 pounds equals zero Recall that the expresson for prce weght n equaton () s: prce weght β + β weght Evaluated at weght 000, the estmated margnal effect of weght on prce n equaton () s: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 4 of 9 pages

25 ECONOMICS 5* -- Stata 0 Tutoral 7 prce estmate of βˆ + ˆ β(000) weght The null and alternatve hypotheses are: H 0 : β + β (000) 0 H : β + β (000) 0 Enter ether of the followng two lncom commands: lncom _b[weght] + *_b[weghtsq]*000 lncom weght + *weghtsq*000 Note that n the second of the above lncom commands varable names are used to represent ther correspondng coeffcent estmates But f you fnd ths notaton confusng, just use the frst form of the above lncom commands Inspect the output generated by the above lncom commands Would you nfer that the margnal effect of weght on prce s dfferent from zero for a car that weghs,000 pounds? Test 8: Test the hypothess that the margnal effect of mpg on prce s zero n regresson equaton () The margnal effect of mpg on prce s obtaned by partally dfferentatng regresson equaton () wth respect to mpg prce mpg β + β 4 mpg Suffcent condtons for prce mpg 0 for all are β 0 and β 4 0 We therefore want to test these two coeffcent excluson restrctons The null and alternatve hypotheses are: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 5 of 9 pages

26 ECONOMICS 5* -- Stata 0 Tutoral 7 H 0 : β 0 and β 4 0 H : β 0 and/or β 4 0 The followng test command computes an F-test of H 0 aganst H Enter the command: test mpg mpgsq Inspect the results generated by ths command State the nference you would draw from ths test Test 9: Use a lncom command to compute the estmated value of the margnal effect of mpg n equaton () for a car that gets 5 mles to the gallon, and to compute a two-tal t-test of the null hypothess that the margnal effect of mpg n equaton () for a car that gets 5 mles to the gallon equals zero Recall that the expresson for prce mpg n equaton () s: prce mpg β + β 4 mpg Evaluated at mpg 5, the estmated margnal effect of mpg on prce s: prce estmate of βˆ + ˆ β4(5) mpg The null and alternatve hypotheses are: H 0 : β + β 4 (5) 0 H : β + β 4 (5) 0 Enter ether (or both) of the followng lncom commands: lncom _b[mpg] + *_b[mpgsq]*5 lncom mpg + *mpgsq*5 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 6 of 9 pages

27 ECONOMICS 5* -- Stata 0 Tutoral 7 What s the estmated value of the margnal effect of mpg on prce for a car that has fuel effcency equal to 5 mles per gallon? Would you nfer that the margnal effect of mpg on prce s dfferent from zero for a car that has fuel effcency equal to 5 mles per gallon? Interacton Varables: Interactons Between Two Contnuous Varables Nature: Interactons between two contnuous varables refer to products of explanatory varables For example, f X j and X k are two contnuous explanatory varables, the nteracton term between them s the product X j X k The product X j X k s ncluded as a regressor n the populaton regresson functon Usage: An nteracton term between two contnuous varables allows the margnal effect of one explanatory varable to be a lnear functon of the other explanatory varable Consder the followng regresson equaton: Y β + β X + β X + β X + β X + β X X + u () Compute the margnal effects on Y of the two explanatory varables X and X n equaton () The margnal effect of X n equaton () s: Y X β + β X + β 5 X a lnear functon of both X and X The margnal effect of X n equaton () s: Y X β + β 4 X + β 5 X a lnear functon of both X and X ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 7 of 9 pages

28 ECONOMICS 5* -- Stata 0 Tutoral 7 Example : polynomals and an nteracton term n two explanatory varables prce β0 + βweght + βmpg + βweght + β4mpg + β5weghtmpg + u () Note: For PRE (), the number of regresson coeffcents s K 6 Regresson equaton () contans two explanatory varables, weght and mpg, but fve regressors weght, weght, mpg, mpg, and weght mpg Margnal Effects of the Explanatory Varables The margnal effect of weght n equaton () s: prce weght β + β weght + β 5 mpg a lnear functon of both weght and mpg The margnal effect of mpg n equaton () s: prce mpg β + β 4 mpg + β 5 weght a lnear functon of both weght and mpg Before estmatng regresson equaton (), t s necessary to create the nteracton term weght mpg ) Enter the command: ( generate wgtmpg weght*mpg Now estmate by OLS regresson equaton () Enter the command: regress prce weght weghtsq mpg mpgsq wgtmpg ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 8 of 9 pages

29 ECONOMICS 5* -- Stata 0 Tutoral 7 Test 0: Test the hypothess that the margnal effect of weght on prce s zero n regresson equaton () The margnal effect of weght on prce n equaton () s prce weght β + β weght + β 5 mpg Suffcent condtons for prce weght 0 for all are β 0 and β 0 and β 5 0 We therefore want to test these three coeffcent excluson restrctons The null and alternatve hypotheses are: H 0 : β 0 and β 0 and β 5 0 (4) H : β 0 and/or β 0 and/or β 5 0 The followng test command computes an F-test of H 0 aganst H n (4) Enter the command: test weght weghtsq wgtmpg Inspect the results generated by ths command State the nference you would draw from ths test Test : Use a lncom command to compute the estmated value of the margnal effect of weght n equaton () for a car that weghs,000 pounds and gets 0 mles to the gallon, and to perform a two-tal t-test of the null hypothess that ths margnal effect of weght n equaton () s equal to zero Recall that the expresson for prce weght n equaton () s: prce weght β + β weght + β 5 mpg Evaluated at weght 000 and mpg 0, the estmated margnal effect of weght on prce s: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 9 of 9 pages

30 ECONOMICS 5* -- Stata 0 Tutoral 7 prce estmate of βˆ ˆ (000) ˆ + β + β5(0) weght The null and alternatve hypotheses are: H 0 : β + β (000) + β (0) 0 5 H : β + β (000) + β (0) 0 5 Enter ether (or both) of the followng lncom commands: lncom _b[weght] + *_b[weghtsq]*000 + _b[wgtmpg]*0 lncom weght + *weghtsq*000 + wgtmpg*0 Inspect the output generated by the above lncom commands Would you nfer that the margnal effect of weght on prce s dfferent from zero for a car that weghs,000 pounds and has fuel effcency equal to 0 mles per gallon? Test : Test the hypothess that the margnal effect of mpg on prce s zero n regresson equaton () The margnal effect of mpg on prce s obtaned by partally dfferentatng regresson equaton () wth respect to mpg prce mpg β + β 4 mpg + β 5 weght Suffcent condtons for prce mpg 0 for all are β 0 and β 4 0 and β 5 0 We therefore want to test these three coeffcent restrctons The null and alternatve hypotheses are: H 0 : β 0 and β 4 0 and β 5 0 (5) H : β 0 and/or β 4 0 and/or β 5 0 The followng test command computes an F-test of H 0 aganst H n (5) Enter the command: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 0 of 9 pages

31 ECONOMICS 5* -- Stata 0 Tutoral 7 test mpg mpgsq wgtmpg Inspect the results generated by ths command State the nference you would draw from ths test Test : Use a lncom command to compute the estmated value of the margnal effect of mpg n equaton () for a car that weghs,000 pounds and gets 0 mles to the gallon, and to perform a two-tal t-test of the null hypothess that the margnal effect of mpg n equaton () for a car that weghs,000 pounds and gets 0 mles to the gallon equals zero Recall that the expresson for prce mpg n equaton () s: prce mpg β + β 4 mpg + β 5 weght Evaluated at weght 000 and mpg 0, the estmated margnal effect of mpg on prce s: prce estmate of βˆ ˆ (0) ˆ + β4 + β5(000) mpg The null and alternatve hypotheses are: H 0 : β + β (0) + β (000) H : β + β (0) + β (000) Enter ether (or both) of the followng lncom commands: lncom _b[mpg] + *_b[mpgsq]*0 + _b[wgtmpg]*000 lncom mpg + *mpgsq*0 + wgtmpg*000 Would you nfer that the margnal effect of mpg on prce s dfferent from zero for a car that weghs,000 pounds and has fuel effcency equal to 0 mles per gallon? What s the estmated value of ths margnal effect of mpg on prce? ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

32 ECONOMICS 5* -- Stata 0 Tutoral 7 Testng the Jont Sgnfcance of All Slope Coeffcent Estmates -- Alternatve Methods A basc test of any estmated sample regresson equaton s a test of the jont sgnfcance of all ts slope coeffcent estmates The reason for the mportance of ths partcular hypothess test s that addresses an mportant queston about the emprcal usefulness and relevance of the postulated model That queston s: Are the regressors n the regresson functon jontly relevant n explanng the varaton of the dependent varable? Do all the regressors together actually help account for the varaton n the dependent varable? Ths secton demonstrates the equvalence of three alternatve methods for computng an F-test of the jont sgnfcance of all the slope coeffcent estmates n a multple lnear regresson model Snce a multple regresson model contans at least two non-constant regressors, the null hypothess for such a jont sgnfcance test specfes two or more coeffcent excluson restrctons An F-test s therefore approprate The Model To llustrate the alternatve methods of performng an F-test of the jont sgnfcance of all slope coeffcent estmates, use the multple lnear regresson model for car prces gven by equaton (): prce β0 + βweght + βmpg + βweght + β4mpg + β5weghtmpg + u () For regresson model (), the total number of regresson coeffcents s K 6, and the number of slope coeffcents s k K 6 5 Re-estmate model () by enterng the followng command: regress prce weght mpg weghtsq mpgsq wgtmpg Alternatvely, you can redsplay the results of the most recently ssued regress command by smply enterng the command: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

33 ECONOMICS 5* -- Stata 0 Tutoral 7 regress The Null and Alternatve Hypotheses H 0 : β j 0 for all j,, 5 β 0 and β 0 and β 5 0; (6) H : β j 0 for j,, 5 β 0 and/or β 0 and/or β 5 0 Method : the ANOVA F-statstc Method uses the ANOVA F-statstc derved from the Analyss-of-Varance table for the OLS sample regresson equaton obtaned by OLS estmaton of equaton () The ANOVA F-statstc can be computed usng two alternatve but equvalent formulas: ESS dfess ESS (K ) R (K ) (7) RSS df RSS (N K) ( R ) (N K) FANOVA RSS where: ESS the Explaned Sum of Squares for the unrestrcted OLS-SRE; RSS the Resdual Sum of Squares for the unrestrcted OLS-SRE; df ESS K the degrees of freedom for ESS; df RSS N K the degrees of freedom for RSS; K the total number of regresson coeffcents n the unrestrcted model; N the number of sample observatons n the estmaton sample R ESS RSS the R-squared for the unrestrcted OLS-SRE TSS TSS The null dstrbuton of the ANOVA F-statstc -- that s, the dstrbuton of F ANOVA under the null hypothess that all K slope coeffcents jontly equal zero s the F[K, N K] F[df ESS, df RSS ] dstrbuton: F ~ F[K, N K] F[df, df ] under H 0 : β j 0 j,, k ANOVA ESS RSS ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page of 9 pages

34 ECONOMICS 5* -- Stata 0 Tutoral 7 Recall from Stata Tutoral that selected results of the regress command are temporarly stored n e( ) functons To compute the ANOVA F-statstc, you wll want to save the values of the terms that enter the formulas for F ANOVA To do ths, enter the followng scalar commands: scalar N e(n) scalar ESS e(mss) scalar dfess e(df_m) scalar RSS e(rss) scalar dfrss e(df_r) scalar TSS ESS + RSS scalar dftss N - scalar lst N ESS dfess RSS dfrss TSS dftss Calculate the sample value of the ANOVA F-statstc usng the formula F ANOVA ESS dfess ESS (K ) (7a) RSS df RSS (N K) RSS Enter the commands: scalar Fanova (ESS/dfESS)/(RSS/dfRSS) scalar lst Fanova ESS RSS dfess dfrss Calculate and dsplay the p-value for the ANOVA F-statstc F ANOVA Enter the command: dsplay Ftal(dfESS, dfrss, Fanova) Calculate and dsplay the 5% and % crtcal values of the F[K, N K] F[df ESS, df RSS ] F[5, 68] dstrbuton Enter the commands: dsplay nvftal(dfess, dfrss, 005) dsplay nvftal(dfess, dfrss, 00) Now calculate the sample value of the ANOVA F-statstc usng the alternatve formula R dfess R (K ) (7b) ( R ) df ( R ) (N K) FANOVA RSS ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 4 of 9 pages

35 ECONOMICS 5* -- Stata 0 Tutoral 7 You wll frst want to generate the value of R, the R-squared for the estmated sample regresson equaton Enter the commands: scalar Rsq ESS/TSS scalar Fanova (Rsq/dfESS)/((-Rsq)/dfRSS) scalar lst Fanova Rsq dfess dfrss Calculate and dsplay the p-value for the ANOVA F-statstc F ANOVA Enter the command: dsplay Ftal(dfESS, dfrss, Fanova) Fnally, compare the sample values of the two ANOVA F-statstcs F ANOVA and F ANOVA Enter the command: scalar lst Fanova Fanova As expected, F ANOVA F ANOVA Method : the General Form F-statstc A second method of computng an F-test of hypothess (6) employs the followng general form of the F-statstc (see Note 7): F 0 (RSS RSS 0 0 (8) RSS ) df (df df ) where: RSS 0 the resdual sum of squares for the restrcted OLS-SRE; RSS the resdual sum of squares for the unrestrcted OLS-SRE; df 0 N K 0 the degrees of freedom for RSS 0, the restrcted RSS; df N K the degrees of freedom for RSS, the unrestrcted RSS; K 0 the number of free regresson coeffcents n the restrcted model; K the number of free regresson coeffcents n the unrestrcted model; df 0 df N K 0 (N K) N K 0 N + K K K 0 ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 5 of 9 pages

36 ECONOMICS 5* -- Stata 0 Tutoral 7 Ths general form F-statstc nvolves the resdual sum of squares for both the unrestrcted OLS sample regresson equaton correspondng to the alternatve hypothess H and the restrcted OLS sample regresson equaton correspondng to the null hypothess H 0 To compute ths general form F-statstc, t s therefore necessary to estmate by OLS both the unrestrcted model mpled the alternatve hypothess H and the restrcted model mpled by the null hypothess H 0 The unrestrcted model mpled the alternatve hypothess H n (6) s smply regresson equaton (): prce β0 + βweght + βmpg + βweght + β4mpg + β5weghtmpg + u () The restrcted model mpled the null hypothess H 0 n (6) s obtaned by settng β 0 and β 0 and β 0 and β 4 0 and β 5 0 n the unrestrcted regresson equaton (): prce β + u 0 (9) The followng seres of commands llustrates the computaton of the general form F-statstc (8) Estmate by OLS the unrestrcted PRE () and retreve the unrestrcted RSS (RSS ) and ts degrees-of-freedom (df ) Enter the commands: regress prce weght mpg weghtsq mpgsq wgtmpg scalar RSS e(rss) scalar df e(df_r) scalar lst RSS df Save for later use the sample value of the F-statstc generated by the above regress command Enter the commands: scalar Fstateq0 e(f) scalar lst Fstateq0 dfess dfrss Estmate by OLS the restrcted PRE (9) and retreve the restrcted RSS (RSS 0 ) and ts degrees-of-freedom (df 0 ) Enter the commands: regress prce ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 6 of 9 pages

37 ECONOMICS 5* -- Stata 0 Tutoral 7 scalar RSS0 e(rss) scalar df0 e(df_r) scalar lst RSS0 df0 To see how to nterpret the OLS estmate of the ntercept coeffcent β 0 n the restrcted regresson equaton (9), compute the sample mean of the dependent varable prce Enter the command: summarze prce ~ Compare the results of ths command wth the restrcted OLS estmate β0 the ntercept coeffcent β 0 n restrcted regresson equaton (9): you wll observe that β ~ 0 prce, the sample mean of the dependent varable prce of Calculate and dsplay the sample value of the F-statstc Enter the commands: scalar F0 ((RSS0 - RSS)/(df0 - df))/(rss/df) scalar lst F0 RSS0 RSS df0 df Calculate the p-value of the calculated F-statstc F 0 Frst create scalars representng the numerator degrees of freedom dfnum df0 df K K0 and the denomnator degrees of freedom dfden df N K for the null dstrbuton of F 0 Enter the commands: scalar dfnum df0 - df scalar dfden df scalar lst dfnum dfden dsplay Ftal(dfnum, dfden, F0) Use the followng commands to compute the crtcal values of the F[dfnum, dfden] F[5, 68]-dstrbuton for sgnfcance levels α 00, 005, 00: dsplay nvftal(5, 68, 00) dsplay nvftal(5, 68, 005) dsplay nvftal(5, 68, 00) Fnally, compare the sample values of the two ANOVA F-statstcs F ANOVA and F ANOVA computed by Method wth the sample value of the F-statstc F 0 computed by Method and the sample value of the F-statstc that s outputted by the regress command for the unrestrcted equaton () Enter the command: ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 7 of 9 pages

38 ECONOMICS 5* -- Stata 0 Tutoral 7 scalar lst Fanova Fanova F0 Fstateq0 As expected, these four F-statstcs all have the same sample value What nference would you draw from these F-statstcs respectng the jont sgnfcance of the slope coeffcent estmates n equaton ()? Method : the test command As you have learned, a sngle test command can be used to perform an F-test of the jont sgnfcance of all slope coeffcent estmates n a lnear regresson model such as equaton () Ths test command produces the same sample value of the F-statstc as do Methods and above, but t uses yet another form of the F- statstc The followng seres of commands llustrates the equvalence of the test command and the other methods Re-estmate regresson equaton () Enter agan the followng regress command: regress prce weght mpg weghtsq mpgsq wgtmpg Now enter the followng test command: test weght mpg weghtsq mpgsq wgtmpg Ths test command performs a jont F-test of the null hypothess H 0 n (6) that all fve slope coeffcents n equaton () are jontly equal to zero aganst the alternatve hypothess H that at least one of these slope coeffcents s not equal to zero Compare the sample value and p-value of the F-statstc computed by ths test command wth those of the F-statstcs computed by Methods and above Enter the commands: scalar lst Fanova Fanova F0 Fstateq0 dsplay Ftal(dfnum, dfden, F0) dsplay Ftal(dfESS, dfrss, Fanova) dsplay Ftal(dfESS, dfrss, Fstateq0) ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 8 of 9 pages

39 ECONOMICS 5* -- Stata 0 Tutoral 7 Preparng to End Your Stata Sesson Before you end your Stata sesson, you should do three thngs Frst, you may want to save the current dataset (although you wll not need t for future tutorals) Enter the followng save command to save the current dataset as Stata-format dataset auto7dta: save auto7 Second, close the command log fle you have been recordng Enter the command: cmdlog close Thrd, close the log fle you have been recordng Enter the command: log close End Your Stata Sesson -- ext To end your Stata sesson, use the ext command Enter the command: ext or ext, clear Cleanng Up and Clearng Out After returnng to Wndows, you should copy all the fles you have used and created durng your Stata sesson to your own dskette These fles wll be found n the Stata workng drectory, whch s usually C:\data on the computers n Dunnng 50, and D:\courses on the computers n MC B There are two fles you wll want to be sure you have: the Stata log fle 5tutoral7log, and the command log fle 5tutoral7txt If you saved the Stata-format data set auto7dta, you wll probably want to take t wth you as well Use the Wndows copy command to copy any fles you want to keep to your own portable electronc storage devce (eg, flash memory stck) n the E:-drve (or to a dskette n the A:- drve) Fnally, as a courtesy to other users of the computng classroom, please delete all the fles you have used or created from the Stata workng drectory ECON 5* -- Wnter 008: Stata 0 Tutoral 7 Page 9 of 9 pages

CHAPTER 14 MORE ABOUT REGRESSION

CHAPTER 14 MORE ABOUT REGRESSION CHAPTER 14 MORE ABOUT REGRESSION We learned n Chapter 5 that often a straght lne descrbes the pattern of a relatonshp between two quanttatve varables. For nstance, n Example 5.1 we explored the relatonshp

More information

Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting

Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting Causal, Explanatory Forecastng Assumes cause-and-effect relatonshp between system nputs and ts output Forecastng wth Regresson Analyss Rchard S. Barr Inputs System Cause + Effect Relatonshp The job of

More information

PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 12

PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 12 14 The Ch-squared dstrbuton PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 1 If a normal varable X, havng mean µ and varance σ, s standardsed, the new varable Z has a mean 0 and varance 1. When ths standardsed

More information

SIMPLE LINEAR CORRELATION

SIMPLE LINEAR CORRELATION SIMPLE LINEAR CORRELATION Smple lnear correlaton s a measure of the degree to whch two varables vary together, or a measure of the ntensty of the assocaton between two varables. Correlaton often s abused.

More information

How To Calculate The Accountng Perod Of Nequalty

How To Calculate The Accountng Perod Of Nequalty Inequalty and The Accountng Perod Quentn Wodon and Shlomo Ytzha World Ban and Hebrew Unversty September Abstract Income nequalty typcally declnes wth the length of tme taen nto account for measurement.

More information

PRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB.

PRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB. PRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB. INDEX 1. Load data usng the Edtor wndow and m-fle 2. Learnng to save results from the Edtor wndow. 3. Computng the Sharpe Rato 4. Obtanng the Treynor Rato

More information

THE METHOD OF LEAST SQUARES THE METHOD OF LEAST SQUARES

THE METHOD OF LEAST SQUARES THE METHOD OF LEAST SQUARES The goal: to measure (determne) an unknown quantty x (the value of a RV X) Realsaton: n results: y 1, y 2,..., y j,..., y n, (the measured values of Y 1, Y 2,..., Y j,..., Y n ) every result s encumbered

More information

NPAR TESTS. One-Sample Chi-Square Test. Cell Specification. Observed Frequencies 1O i 6. Expected Frequencies 1EXP i 6

NPAR TESTS. One-Sample Chi-Square Test. Cell Specification. Observed Frequencies 1O i 6. Expected Frequencies 1EXP i 6 PAR TESTS If a WEIGHT varable s specfed, t s used to replcate a case as many tmes as ndcated by the weght value rounded to the nearest nteger. If the workspace requrements are exceeded and samplng has

More information

1. Measuring association using correlation and regression

1. Measuring association using correlation and regression How to measure assocaton I: Correlaton. 1. Measurng assocaton usng correlaton and regresson We often would lke to know how one varable, such as a mother's weght, s related to another varable, such as a

More information

4 Hypothesis testing in the multiple regression model

4 Hypothesis testing in the multiple regression model 4 Hypothess testng n the multple regresson model Ezequel Urel Unversdad de Valenca Verson: 9-13 4.1 Hypothess testng: an overvew 1 4.1.1 Formulaton of the null hypothess and the alternatve hypothess 4.1.

More information

Economic Interpretation of Regression. Theory and Applications

Economic Interpretation of Regression. Theory and Applications Economc Interpretaton of Regresson Theor and Applcatons Classcal and Baesan Econometrc Methods Applcaton of mathematcal statstcs to economc data for emprcal support Economc theor postulates a qualtatve

More information

5 Multiple regression analysis with qualitative information

5 Multiple regression analysis with qualitative information 5 Multple regresson analyss wth qualtatve nformaton Ezequel Urel Unversty of Valenca Verson: 9-13 5.1 Introducton of qualtatve nformaton n econometrc models. 1 5. A sngle dummy ndependent varable 5.3 Multple

More information

CHAPTER 5 RELATIONSHIPS BETWEEN QUANTITATIVE VARIABLES

CHAPTER 5 RELATIONSHIPS BETWEEN QUANTITATIVE VARIABLES CHAPTER 5 RELATIONSHIPS BETWEEN QUANTITATIVE VARIABLES In ths chapter, we wll learn how to descrbe the relatonshp between two quanttatve varables. Remember (from Chapter 2) that the terms quanttatve varable

More information

Online Appendix for Forecasting the Equity Risk Premium: The Role of Technical Indicators

Online Appendix for Forecasting the Equity Risk Premium: The Role of Technical Indicators Onlne Appendx for Forecastng the Equty Rsk Premum: The Role of Techncal Indcators Chrstopher J. Neely Federal Reserve Bank of St. Lous neely@stls.frb.org Davd E. Rapach Sant Lous Unversty rapachde@slu.edu

More information

Texas Instruments 30X IIS Calculator

Texas Instruments 30X IIS Calculator Texas Instruments 30X IIS Calculator Keystrokes for the TI-30X IIS are shown for a few topcs n whch keystrokes are unque. Start by readng the Quk Start secton. Then, before begnnng a specfc unt of the

More information

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ). REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or

More information

Can Auto Liability Insurance Purchases Signal Risk Attitude?

Can Auto Liability Insurance Purchases Signal Risk Attitude? Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang

More information

Institute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic

Institute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic Lagrange Multplers as Quanttatve Indcators n Economcs Ivan Mezník Insttute of Informatcs, Faculty of Busness and Management, Brno Unversty of TechnologCzech Republc Abstract The quanttatve role of Lagrange

More information

1 Example 1: Axis-aligned rectangles

1 Example 1: Axis-aligned rectangles COS 511: Theoretcal Machne Learnng Lecturer: Rob Schapre Lecture # 6 Scrbe: Aaron Schld February 21, 2013 Last class, we dscussed an analogue for Occam s Razor for nfnte hypothess spaces that, n conjuncton

More information

8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by

8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by 6 CHAPTER 8 COMPLEX VECTOR SPACES 5. Fnd the kernel of the lnear transformaton gven n Exercse 5. In Exercses 55 and 56, fnd the mage of v, for the ndcated composton, where and are gven by the followng

More information

Risk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008

Risk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008 Rsk-based Fatgue Estmate of Deep Water Rsers -- Course Project for EM388F: Fracture Mechancs, Sprng 2008 Chen Sh Department of Cvl, Archtectural, and Envronmental Engneerng The Unversty of Texas at Austn

More information

Calibration and Linear Regression Analysis: A Self-Guided Tutorial

Calibration and Linear Regression Analysis: A Self-Guided Tutorial Calbraton and Lnear Regresson Analyss: A Self-Guded Tutoral Part The Calbraton Curve, Correlaton Coeffcent and Confdence Lmts CHM314 Instrumental Analyss Department of Chemstry, Unversty of Toronto Dr.

More information

Calculation of Sampling Weights

Calculation of Sampling Weights Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample

More information

Section 5.4 Annuities, Present Value, and Amortization

Section 5.4 Annuities, Present Value, and Amortization Secton 5.4 Annutes, Present Value, and Amortzaton Present Value In Secton 5.2, we saw that the present value of A dollars at nterest rate per perod for n perods s the amount that must be deposted today

More information

1 De nitions and Censoring

1 De nitions and Censoring De ntons and Censorng. Survval Analyss We begn by consderng smple analyses but we wll lead up to and take a look at regresson on explanatory factors., as n lnear regresson part A. The mportant d erence

More information

STATISTICAL DATA ANALYSIS IN EXCEL

STATISTICAL DATA ANALYSIS IN EXCEL Mcroarray Center STATISTICAL DATA ANALYSIS IN EXCEL Lecture 6 Some Advanced Topcs Dr. Petr Nazarov 14-01-013 petr.nazarov@crp-sante.lu Statstcal data analyss n Ecel. 6. Some advanced topcs Correcton for

More information

An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services

An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services An Evaluaton of the Extended Logstc, Smple Logstc, and Gompertz Models for Forecastng Short Lfecycle Products and Servces Charles V. Trappey a,1, Hsn-yng Wu b a Professor (Management Scence), Natonal Chao

More information

Recurrence. 1 Definitions and main statements

Recurrence. 1 Definitions and main statements Recurrence 1 Defntons and man statements Let X n, n = 0, 1, 2,... be a MC wth the state space S = (1, 2,...), transton probabltes p j = P {X n+1 = j X n = }, and the transton matrx P = (p j ),j S def.

More information

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK Sample Stablty Protocol Background The Cholesterol Reference Method Laboratory Network (CRMLN) developed certfcaton protocols for total cholesterol, HDL

More information

What is Candidate Sampling

What is Candidate Sampling What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble

More information

Regression Models for a Binary Response Using EXCEL and JMP

Regression Models for a Binary Response Using EXCEL and JMP SEMATECH 997 Statstcal Methods Symposum Austn Regresson Models for a Bnary Response Usng EXCEL and JMP Davd C. Trndade, Ph.D. STAT-TECH Consultng and Tranng n Appled Statstcs San Jose, CA Topcs Practcal

More information

Online Appendix Supplemental Material for Market Microstructure Invariance: Empirical Hypotheses

Online Appendix Supplemental Material for Market Microstructure Invariance: Empirical Hypotheses Onlne Appendx Supplemental Materal for Market Mcrostructure Invarance: Emprcal Hypotheses Albert S. Kyle Unversty of Maryland akyle@rhsmth.umd.edu Anna A. Obzhaeva New Economc School aobzhaeva@nes.ru Table

More information

Using Series to Analyze Financial Situations: Present Value

Using Series to Analyze Financial Situations: Present Value 2.8 Usng Seres to Analyze Fnancal Stuatons: Present Value In the prevous secton, you learned how to calculate the amount, or future value, of an ordnary smple annuty. The amount s the sum of the accumulated

More information

An Alternative Way to Measure Private Equity Performance

An Alternative Way to Measure Private Equity Performance An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate

More information

Exhaustive Regression. An Exploration of Regression-Based Data Mining Techniques Using Super Computation

Exhaustive Regression. An Exploration of Regression-Based Data Mining Techniques Using Super Computation Exhaustve Regresson An Exploraton of Regresson-Based Data Mnng Technques Usng Super Computaton Antony Daves, Ph.D. Assocate Professor of Economcs Duquesne Unversty Pttsburgh, PA 58 Research Fellow The

More information

Lecture 3: Force of Interest, Real Interest Rate, Annuity

Lecture 3: Force of Interest, Real Interest Rate, Annuity Lecture 3: Force of Interest, Real Interest Rate, Annuty Goals: Study contnuous compoundng and force of nterest Dscuss real nterest rate Learn annuty-mmedate, and ts present value Study annuty-due, and

More information

DEFINING %COMPLETE IN MICROSOFT PROJECT

DEFINING %COMPLETE IN MICROSOFT PROJECT CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,

More information

HOUSEHOLDS DEBT BURDEN: AN ANALYSIS BASED ON MICROECONOMIC DATA*

HOUSEHOLDS DEBT BURDEN: AN ANALYSIS BASED ON MICROECONOMIC DATA* HOUSEHOLDS DEBT BURDEN: AN ANALYSIS BASED ON MICROECONOMIC DATA* Luísa Farnha** 1. INTRODUCTION The rapd growth n Portuguese households ndebtedness n the past few years ncreased the concerns that debt

More information

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..

More information

BERNSTEIN POLYNOMIALS

BERNSTEIN POLYNOMIALS On-Lne Geometrc Modelng Notes BERNSTEIN POLYNOMIALS Kenneth I. Joy Vsualzaton and Graphcs Research Group Department of Computer Scence Unversty of Calforna, Davs Overvew Polynomals are ncredbly useful

More information

Faraday's Law of Induction

Faraday's Law of Induction Introducton Faraday's Law o Inducton In ths lab, you wll study Faraday's Law o nducton usng a wand wth col whch swngs through a magnetc eld. You wll also examne converson o mechanc energy nto electrc energy

More information

Is There A Tradeoff between Employer-Provided Health Insurance and Wages?

Is There A Tradeoff between Employer-Provided Health Insurance and Wages? Is There A Tradeoff between Employer-Provded Health Insurance and Wages? Lye Zhu, Southern Methodst Unversty October 2005 Abstract Though most of the lterature n health nsurance and the labor market assumes

More information

Addendum to: Importing Skill-Biased Technology

Addendum to: Importing Skill-Biased Technology Addendum to: Importng Skll-Based Technology Arel Bursten UCLA and NBER Javer Cravno UCLA August 202 Jonathan Vogel Columba and NBER Abstract Ths Addendum derves the results dscussed n secton 3.3 of our

More information

Marginal Benefit Incidence Analysis Using a Single Cross-section of Data. Mohamed Ihsan Ajwad and Quentin Wodon 1. World Bank.

Marginal Benefit Incidence Analysis Using a Single Cross-section of Data. Mohamed Ihsan Ajwad and Quentin Wodon 1. World Bank. Margnal Beneft Incdence Analyss Usng a Sngle Cross-secton of Data Mohamed Ihsan Ajwad and uentn Wodon World Bank August 200 Abstract In a recent paper, Lanjouw and Ravallon proposed an attractve and smple

More information

Chapter XX More advanced approaches to the analysis of survey data. Gad Nathan Hebrew University Jerusalem, Israel. Abstract

Chapter XX More advanced approaches to the analysis of survey data. Gad Nathan Hebrew University Jerusalem, Israel. Abstract Household Sample Surveys n Developng and Transton Countres Chapter More advanced approaches to the analyss of survey data Gad Nathan Hebrew Unversty Jerusalem, Israel Abstract In the present chapter, we

More information

Level Annuities with Payments Less Frequent than Each Interest Period

Level Annuities with Payments Less Frequent than Each Interest Period Level Annutes wth Payments Less Frequent than Each Interest Perod 1 Annuty-mmedate 2 Annuty-due Level Annutes wth Payments Less Frequent than Each Interest Perod 1 Annuty-mmedate 2 Annuty-due Symoblc approach

More information

International University of Japan Public Management & Policy Analysis Program

International University of Japan Public Management & Policy Analysis Program Internatonal Unversty of Japan Publc Management & Polcy Analyss Program Practcal Gudes To Panel Data Modelng: A Step by Step Analyss Usng Stata * Hun Myoung Park, Ph.D. kucc65@uj.ac.jp 1. Introducton.

More information

Meta-Analysis of Hazard Ratios

Meta-Analysis of Hazard Ratios NCSS Statstcal Softare Chapter 458 Meta-Analyss of Hazard Ratos Introducton Ths module performs a meta-analyss on a set of to-group, tme to event (survval), studes n hch some data may be censored. These

More information

The Application of Fractional Brownian Motion in Option Pricing

The Application of Fractional Brownian Motion in Option Pricing Vol. 0, No. (05), pp. 73-8 http://dx.do.org/0.457/jmue.05.0..6 The Applcaton of Fractonal Brownan Moton n Opton Prcng Qng-xn Zhou School of Basc Scence,arbn Unversty of Commerce,arbn zhouqngxn98@6.com

More information

n + d + q = 24 and.05n +.1d +.25q = 2 { n + d + q = 24 (3) n + 2d + 5q = 40 (2)

n + d + q = 24 and.05n +.1d +.25q = 2 { n + d + q = 24 (3) n + 2d + 5q = 40 (2) MATH 16T Exam 1 : Part I (In-Class) Solutons 1. (0 pts) A pggy bank contans 4 cons, all of whch are nckels (5 ), dmes (10 ) or quarters (5 ). The pggy bank also contans a con of each denomnaton. The total

More information

PRIVATE SCHOOL CHOICE: THE EFFECTS OF RELIGIOUS AFFILIATION AND PARTICIPATION

PRIVATE SCHOOL CHOICE: THE EFFECTS OF RELIGIOUS AFFILIATION AND PARTICIPATION PRIVATE SCHOOL CHOICE: THE EFFECTS OF RELIIOUS AFFILIATION AND PARTICIPATION Danny Cohen-Zada Department of Economcs, Ben-uron Unversty, Beer-Sheva 84105, Israel Wllam Sander Department of Economcs, DePaul

More information

v a 1 b 1 i, a 2 b 2 i,..., a n b n i.

v a 1 b 1 i, a 2 b 2 i,..., a n b n i. SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 455 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces we have studed thus far n the text are real vector spaces snce the scalars are

More information

Marginal Returns to Education For Teachers

Marginal Returns to Education For Teachers The Onlne Journal of New Horzons n Educaton Volume 4, Issue 3 MargnalReturnstoEducatonForTeachers RamleeIsmal,MarnahAwang ABSTRACT FacultyofManagementand Economcs UnverstPenddkanSultan Idrs ramlee@fpe.ups.edu.my

More information

The Mathematical Derivation of Least Squares

The Mathematical Derivation of Least Squares Pscholog 885 Prof. Federco The Mathematcal Dervaton of Least Squares Back when the powers that e forced ou to learn matr algera and calculus, I et ou all asked ourself the age-old queston: When the hell

More information

Canon NTSC Help Desk Documentation

Canon NTSC Help Desk Documentation Canon NTSC Help Desk Documentaton READ THIS BEFORE PROCEEDING Before revewng ths documentaton, Canon Busness Solutons, Inc. ( CBS ) hereby refers you, the customer or customer s representatve or agent

More information

The Effects of Tax Rate Changes on Tax Bases and the Marginal Cost of Public Funds for Canadian Provincial Governments

The Effects of Tax Rate Changes on Tax Bases and the Marginal Cost of Public Funds for Canadian Provincial Governments The Effects of Tax Rate Changes on Tax Bases and the Margnal Cost of Publc Funds for Canadan Provncal Governments Bev Dahlby a and Ergete Ferede b a Department of Economcs, Unversty of Alberta, Edmonton,

More information

Extending Probabilistic Dynamic Epistemic Logic

Extending Probabilistic Dynamic Epistemic Logic Extendng Probablstc Dynamc Epstemc Logc Joshua Sack May 29, 2008 Probablty Space Defnton A probablty space s a tuple (S, A, µ), where 1 S s a set called the sample space. 2 A P(S) s a σ-algebra: a set

More information

Mean Molecular Weight

Mean Molecular Weight Mean Molecular Weght The thermodynamc relatons between P, ρ, and T, as well as the calculaton of stellar opacty requres knowledge of the system s mean molecular weght defned as the mass per unt mole of

More information

Evidence of Adverse Selection in Automobile Insurance Markets

Evidence of Adverse Selection in Automobile Insurance Markets Evdence of Adverse Selecton n Automoble Insurance Markets by Georges Donne, Chrstan Gouréroux and Charles Vanasse Workng Paper 98-09 Aprl 1998 ISSN : 106-330 Ths research was fnanced by CREST and FFSA,

More information

Question 2: What is the variance and standard deviation of a dataset?

Question 2: What is the variance and standard deviation of a dataset? Queston 2: What s the varance and standard devaton of a dataset? The varance of the data uses all of the data to compute a measure of the spread n the data. The varance may be computed for a sample of

More information

Solution: Let i = 10% and d = 5%. By definition, the respective forces of interest on funds A and B are. i 1 + it. S A (t) = d (1 dt) 2 1. = d 1 dt.

Solution: Let i = 10% and d = 5%. By definition, the respective forces of interest on funds A and B are. i 1 + it. S A (t) = d (1 dt) 2 1. = d 1 dt. Chapter 9 Revew problems 9.1 Interest rate measurement Example 9.1. Fund A accumulates at a smple nterest rate of 10%. Fund B accumulates at a smple dscount rate of 5%. Fnd the pont n tme at whch the forces

More information

SPEE Recommended Evaluation Practice #6 Definition of Decline Curve Parameters Background:

SPEE Recommended Evaluation Practice #6 Definition of Decline Curve Parameters Background: SPEE Recommended Evaluaton Practce #6 efnton of eclne Curve Parameters Background: The producton hstores of ol and gas wells can be analyzed to estmate reserves and future ol and gas producton rates and

More information

World currency options market efficiency

World currency options market efficiency Arful Hoque (Australa) World optons market effcency Abstract The World Currency Optons (WCO) maket began tradng n July 2007 on the Phladelpha Stock Exchange (PHLX) wth the new features. These optons are

More information

An Analysis of the relationship between WTI term structure and oil market fundamentals in 2002-2009

An Analysis of the relationship between WTI term structure and oil market fundamentals in 2002-2009 MPRA Munch Personal RePEc Archve An Analyss of the relatonshp between WTI term structure and ol market fundamentals n 00-009 Mleno Cavalcante Petrobras S.A., Unversdade de Fortaleza. August 00 Onlne at

More information

A machine vision approach for detecting and inspecting circular parts

A machine vision approach for detecting and inspecting circular parts A machne vson approach for detectng and nspectng crcular parts Du-Mng Tsa Machne Vson Lab. Department of Industral Engneerng and Management Yuan-Ze Unversty, Chung-L, Tawan, R.O.C. E-mal: edmtsa@saturn.yzu.edu.tw

More information

Chapter 7: Answers to Questions and Problems

Chapter 7: Answers to Questions and Problems 19. Based on the nformaton contaned n Table 7-3 of the text, the food and apparel ndustres are most compettve and therefore probably represent the best match for the expertse of these managers. Chapter

More information

UK Letter Mail Demand: a Content Based Time Series Analysis using Overlapping Market Survey Statistical Techniques

UK Letter Mail Demand: a Content Based Time Series Analysis using Overlapping Market Survey Statistical Techniques 10-170 Research Group: Econometrcs and Statstcs 2010 UK Letter Mal Demand: a Content Based Tme Seres nalyss usng Overlappng Market Survey Statstcal Technques CTHERINE CZLS, JEN-PIERRE FLORENS, LETICI VERUETE-MCKY,

More information

+ + + - - This circuit than can be reduced to a planar circuit

+ + + - - This circuit than can be reduced to a planar circuit MeshCurrent Method The meshcurrent s analog of the nodeoltage method. We sole for a new set of arables, mesh currents, that automatcally satsfy KCLs. As such, meshcurrent method reduces crcut soluton to

More information

Forecasting the Direction and Strength of Stock Market Movement

Forecasting the Direction and Strength of Stock Market Movement Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems

More information

Ring structure of splines on triangulations

Ring structure of splines on triangulations www.oeaw.ac.at Rng structure of splnes on trangulatons N. Vllamzar RICAM-Report 2014-48 www.rcam.oeaw.ac.at RING STRUCTURE OF SPLINES ON TRIANGULATIONS NELLY VILLAMIZAR Introducton For a trangulated regon

More information

The Cross Section of Foreign Currency Risk Premia and Consumption Growth Risk

The Cross Section of Foreign Currency Risk Premia and Consumption Growth Risk The Cross Secton of Foregn Currency Rsk Prema and Consumpton Growth Rsk By HANNO LUSTIG AND ADRIEN VERDELHAN* Aggregate consumpton growth rsk explans why low nterest rate currences do not apprecate as

More information

Diagnostic Tests of Cross Section Independence for Nonlinear Panel Data Models

Diagnostic Tests of Cross Section Independence for Nonlinear Panel Data Models DISCUSSION PAPER SERIES IZA DP No. 2756 Dagnostc ests of Cross Secton Independence for Nonlnear Panel Data Models Cheng Hsao M. Hashem Pesaran Andreas Pck Aprl 2007 Forschungsnsttut zur Zukunft der Arbet

More information

Statistical algorithms in Review Manager 5

Statistical algorithms in Review Manager 5 Statstcal algorthms n Reve Manager 5 Jonathan J Deeks and Julan PT Hggns on behalf of the Statstcal Methods Group of The Cochrane Collaboraton August 00 Data structure Consder a meta-analyss of k studes

More information

YIELD CURVE FITTING 2.0 Constructing Bond and Money Market Yield Curves using Cubic B-Spline and Natural Cubic Spline Methodology.

YIELD CURVE FITTING 2.0 Constructing Bond and Money Market Yield Curves using Cubic B-Spline and Natural Cubic Spline Methodology. YIELD CURVE FITTING 2.0 Constructng Bond and Money Market Yeld Curves usng Cubc B-Splne and Natural Cubc Splne Methodology Users Manual YIELD CURVE FITTING 2.0 Users Manual Authors: Zhuosh Lu, Moorad Choudhry

More information

Characterization of Assembly. Variation Analysis Methods. A Thesis. Presented to the. Department of Mechanical Engineering. Brigham Young University

Characterization of Assembly. Variation Analysis Methods. A Thesis. Presented to the. Department of Mechanical Engineering. Brigham Young University Characterzaton of Assembly Varaton Analyss Methods A Thess Presented to the Department of Mechancal Engneerng Brgham Young Unversty In Partal Fulfllment of the Requrements for the Degree Master of Scence

More information

Answer: A). There is a flatter IS curve in the high MPC economy. Original LM LM after increase in M. IS curve for low MPC economy

Answer: A). There is a flatter IS curve in the high MPC economy. Original LM LM after increase in M. IS curve for low MPC economy 4.02 Quz Solutons Fall 2004 Multple-Choce Questons (30/00 ponts) Please, crcle the correct answer for each of the followng 0 multple-choce questons. For each queston, only one of the answers s correct.

More information

Statistical Methods to Develop Rating Models

Statistical Methods to Develop Rating Models Statstcal Methods to Develop Ratng Models [Evelyn Hayden and Danel Porath, Österrechsche Natonalbank and Unversty of Appled Scences at Manz] Source: The Basel II Rsk Parameters Estmaton, Valdaton, and

More information

Criminal Justice System on Crime *

Criminal Justice System on Crime * On the Impact of the NSW Crmnal Justce System on Crme * Dr Vasls Sarafds, Dscplne of Operatons Management and Econometrcs Unversty of Sydney * Ths presentaton s based on jont work wth Rchard Kelaher 1

More information

7 ANALYSIS OF VARIANCE (ANOVA)

7 ANALYSIS OF VARIANCE (ANOVA) 7 ANALYSIS OF VARIANCE (ANOVA) Chapter 7 Analyss of Varance (Anova) Objectves After studyng ths chapter you should apprecate the need for analysng data from more than two samples; understand the underlyng

More information

Latent Class Regression. Statistics for Psychosocial Research II: Structural Models December 4 and 6, 2006

Latent Class Regression. Statistics for Psychosocial Research II: Structural Models December 4 and 6, 2006 Latent Class Regresson Statstcs for Psychosocal Research II: Structural Models December 4 and 6, 2006 Latent Class Regresson (LCR) What s t and when do we use t? Recall the standard latent class model

More information

total A A reag total A A r eag

total A A reag total A A r eag hapter 5 Standardzng nalytcal Methods hapter Overvew 5 nalytcal Standards 5B albratng the Sgnal (S total ) 5 Determnng the Senstvty (k ) 5D Lnear Regresson and albraton urves 5E ompensatng for the Reagent

More information

Evidence of the unspanned stochastic volatility in crude-oil market

Evidence of the unspanned stochastic volatility in crude-oil market The Academy of Economc Studes The Faculty of Fnance, Insurance, Bankng and Stock Echange Doctoral School of Fnance and Bankng (DOFIN) Dssertaton Paper Evdence of the unspanned stochastc volatlty n crude-ol

More information

We are now ready to answer the question: What are the possible cardinalities for finite fields?

We are now ready to answer the question: What are the possible cardinalities for finite fields? Chapter 3 Fnte felds We have seen, n the prevous chapters, some examples of fnte felds. For example, the resdue class rng Z/pZ (when p s a prme) forms a feld wth p elements whch may be dentfed wth the

More information

Hedging Interest-Rate Risk with Duration

Hedging Interest-Rate Risk with Duration FIXED-INCOME SECURITIES Chapter 5 Hedgng Interest-Rate Rsk wth Duraton Outlne Prcng and Hedgng Prcng certan cash-flows Interest rate rsk Hedgng prncples Duraton-Based Hedgng Technques Defnton of duraton

More information

Luby s Alg. for Maximal Independent Sets using Pairwise Independence

Luby s Alg. for Maximal Independent Sets using Pairwise Independence Lecture Notes for Randomzed Algorthms Luby s Alg. for Maxmal Independent Sets usng Parwse Independence Last Updated by Erc Vgoda on February, 006 8. Maxmal Independent Sets For a graph G = (V, E), an ndependent

More information

Prediction of Disability Frequencies in Life Insurance

Prediction of Disability Frequencies in Life Insurance Predcton of Dsablty Frequences n Lfe Insurance Bernhard Köng Fran Weber Maro V. Wüthrch October 28, 2011 Abstract For the predcton of dsablty frequences, not only the observed, but also the ncurred but

More information

Project Networks With Mixed-Time Constraints

Project Networks With Mixed-Time Constraints Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa

More information

Gender differences in revealed risk taking: evidence from mutual fund investors

Gender differences in revealed risk taking: evidence from mutual fund investors Economcs Letters 76 (2002) 151 158 www.elsever.com/ locate/ econbase Gender dfferences n revealed rsk takng: evdence from mutual fund nvestors a b c, * Peggy D. Dwyer, James H. Glkeson, John A. Lst a Unversty

More information

Evaluating credit risk models: A critique and a new proposal

Evaluating credit risk models: A critique and a new proposal Evaluatng credt rsk models: A crtque and a new proposal Hergen Frerchs* Gunter Löffler Unversty of Frankfurt (Man) February 14, 2001 Abstract Evaluatng the qualty of credt portfolo rsk models s an mportant

More information

WORKING PAPER. C.D. Howe Institute. The Effects of Tax Rate Changes on Tax Bases and the Marginal Cost of Public Funds for Provincial Governments

WORKING PAPER. C.D. Howe Institute. The Effects of Tax Rate Changes on Tax Bases and the Marginal Cost of Public Funds for Provincial Governments MARCH 211 C.D. Howe Insttute WORKING PAPER FISCAL AND TAX COMPETITIVENESS The Effects of Tax Rate Changes on Tax Bases and the Margnal Cost of Publc Funds for Provncal Governments Bev Dahlby Ergete Ferede

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Max Wellng Department of Computer Scence Unversty of Toronto 10 Kng s College Road Toronto, M5S 3G5 Canada wellng@cs.toronto.edu Abstract Ths s a note to explan support vector machnes.

More information

How To Find The Dsablty Frequency Of A Clam

How To Find The Dsablty Frequency Of A Clam 1 Predcton of Dsablty Frequences n Lfe Insurance Bernhard Köng 1, Fran Weber 1, Maro V. Wüthrch 2 Abstract: For the predcton of dsablty frequences, not only the observed, but also the ncurred but not yet

More information

Joe Pimbley, unpublished, 2005. Yield Curve Calculations

Joe Pimbley, unpublished, 2005. Yield Curve Calculations Joe Pmbley, unpublshed, 005. Yeld Curve Calculatons Background: Everythng s dscount factors Yeld curve calculatons nclude valuaton of forward rate agreements (FRAs), swaps, nterest rate optons, and forward

More information

Linear Circuits Analysis. Superposition, Thevenin /Norton Equivalent circuits

Linear Circuits Analysis. Superposition, Thevenin /Norton Equivalent circuits Lnear Crcuts Analyss. Superposton, Theenn /Norton Equalent crcuts So far we hae explored tmendependent (resste) elements that are also lnear. A tmendependent elements s one for whch we can plot an / cure.

More information

Production. 2. Y is closed A set is closed if it contains its boundary. We need this for the solution existence in the profit maximization problem.

Production. 2. Y is closed A set is closed if it contains its boundary. We need this for the solution existence in the profit maximization problem. Producer Theory Producton ASSUMPTION 2.1 Propertes of the Producton Set The producton set Y satsfes the followng propertes 1. Y s non-empty If Y s empty, we have nothng to talk about 2. Y s closed A set

More information

ECONOMICS 351* -- Stata 10 Tutorial 2. Stata 10 Tutorial 2

ECONOMICS 351* -- Stata 10 Tutorial 2. Stata 10 Tutorial 2 Stata 10 Tutorial 2 TOPIC: Introduction to Selected Stata Commands DATA: auto1.dta (the Stata-format data file you created in Stata Tutorial 1) or auto1.raw (the original text-format data file) TASKS:

More information

VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) 2127472, Fax: (370-5) 276 1380, Email: info@teltonika.

VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) 2127472, Fax: (370-5) 276 1380, Email: info@teltonika. VRT012 User s gude V0.1 Thank you for purchasng our product. We hope ths user-frendly devce wll be helpful n realsng your deas and brngng comfort to your lfe. Please take few mnutes to read ths manual

More information

Lecture 2: Single Layer Perceptrons Kevin Swingler

Lecture 2: Single Layer Perceptrons Kevin Swingler Lecture 2: Sngle Layer Perceptrons Kevn Sngler kms@cs.str.ac.uk Recap: McCulloch-Ptts Neuron Ths vastly smplfed model of real neurons s also knon as a Threshold Logc Unt: W 2 A Y 3 n W n. A set of synapses

More information

Lecture 14: Implementing CAPM

Lecture 14: Implementing CAPM Lecture 14: Implementng CAPM Queston: So, how do I apply the CAPM? Current readng: Brealey and Myers, Chapter 9 Reader, Chapter 15 M. Spegel and R. Stanton, 2000 1 Key Results So Far All nvestors should

More information

Testing The Torah Code Hypothesis: The Experimental Protocol

Testing The Torah Code Hypothesis: The Experimental Protocol Testng The Torah Code Hypothess: The Expermental Protocol Robert M. Haralck Computer Scence, Graduate Center Cty Unversty of New York 365 Ffth Avenue New York, NY 10016 Abstract Ths s the second part of

More information