Estimating income equity in social health insurance system

Size: px
Start display at page:

Download "Estimating income equity in social health insurance system"

Transcription

1 Centre for Economc and Fnancal Research at New Economc School Aprl 01 Estmatng ncome equty n socal health nsurance system Galna Besstremyannaya Workng Paper No 17 CEFIR /NES Workng Paper seres

2 Estmatng ncome equty n socal health nsurance system Besstremyannaya Galna 1 Center for Economc and Fnancal Research at New Economc School Aprl 01 Abstract The paper measures horzontal equty n health care access and utlzaton n Japan by estmatng the coeffcents for ncome groups n a mult-part model whch dstngushes between non-users of health care, the users of npatent and outpatent care. To account for consumer unobservable characterstcs, we apply a latent class approach. We address a retransformaton problem of logged health care expendture, usng generalzed lnear models. Our sample s the 009 data for 4,0 adult consumers (Japan Household Panel Survey). The coeffcents for ncome groups are nsgnfcant both n the bnary choce models for npatent/outpatent health care use, and n the models for health care expendture. Consumers separate nto two latent classes n the generalzed lnear model for outpatent health care expendture. Although the results reveal horzontal equty n health care access and utlzaton n Japan, horzontal nequty remans n health nsurance premums and the prevalence of catastrophc coverage. Keywords: health care demand, equty, ncome elastcty, generalzed lnear models, latent class, two-part model, four-part model, socal health nsurance JEL codes: I10, I18, G, R 1 Center for Economc and Fnancal Research at New Economc School, Offce 9, Nakhmovsky prospect, 47, Moscow, Emal: gbesstre (at) cefr.ru. I am ndebted to Ruben Enkolopov (New Economc School) for stmulatng dscusson and kn comments on earler versons of the paper. I apprecate generous help by Serge Golovan (New Economc School). I thank Dmtry Shapro (Unversty of North Carolna Charlotte) and Jaak Smm (Tokyo Insttute of Technology) for moral support. The cooperaton of Keo Unversty Jont Research Center for Panel Studes (Tokyo) n sharng the mcrodata of Japan Household Panel Survey (wave 1, 009) s gratefully acknowledged.

3 1. Introducton Guaranteeng equty for the poor s a major challenge for health care systems n developed countres. Overall, equty s an ethcal ssue related to the judgments about health care accessblty. At the same tme, an economc concept of horzontal equty deals wth an equal treatment for equal need (Wagstaff and van Doorslaer, 1991a; Culyer and Wagstaff, 1993) and means that persons n equal need of medcal care should receve the same treatment, rrespectve of whether they happen to be poor or rch (Wagstaff and van Doorslaer, 1991b). In practcal terms, there s a general agreement about strvng for mnmal varaton of [health care] use wth ncome (Newhouse et al., 1981) and ensurng equty for the poor (Wagstaff and van Doorslaer, 000b; Cutler, 00). Accordng to theoretcal predctons, a well-desgned socal health nsurance system may provde an equtable redstrbuton of medcal care between the rch and the poor (Zwefel and Breyer, 006). However, the actual performance of socal health nsurance systems wth respect to guaranteeng equty for the poor s an ultmately emprcal queston (Hurley, 000; van Doorslaer et al., 004; Rannan-Elya and Somanathan, 006; Wagstaff, 010). The most prevalent method for analyzng ncome equty measures coeffcents for ncome groups n the equaton for health care utlzaton, wth equalty of the coeffcents nterpreted as zero nequty (Wagstaff and van Doorslaer, 000a; Jones and Wldman, 008). The regresson method should also be regarded as the most general. Indeed, the non-rejecton of the null hypothess of equalty of coeffcents for ncome groups provdes a suffcent condton for zero nequty n terms of an alternatve approach, whch measures concentraton ndces (Wagstaff and van Doorslaer, 1991b; Wagstaff and van Doorslaer, 000a; Wagstaff and van Doorslaer, 000b). Regresson method commonly regards the state of health as the major covarate that should have a sgnfcant estmated coeffcent (.e., need explanatory varable). 3 However, owng to lmtatons of most mcrodata surveys, qualtatve parameters related to the state of health (e.g., self-assessed health) may mal fal to fully capture ndvdual s demand for health care. Therefore, ncorporatng consumer s unobservable characterstcs whch nfluence the decson about health care use, as well as the amount As s defned n The Dctonary of Health Economcs, equty relates n general to ethcal judgments about the farness of ncome and wealth dstrbutons, cost and beneft dstrbutons, accessblty of health servces, exposure to health-threatenng hazards (Culyer, 005). 3 Indeed, the healthy and the sck have dfferent ncome elastcty of health care expendture (Nyman, 006). 3

4 of health care purchased, s essental for rasng the precson of the estmatons of health care demand. The purpose of ths paper s to estmate ncome equty n health care access and utlzaton n Japanese socal health nsurance system. Despte recent concerns about the poor n Japan, the fndngs on ncome effect for health care demand are lmted and mxed. Income effect s nsgnfcant accordng to the results of some studes (Senoo, 1985; Sawano, 001; I and Ohkusa, 00a; Kawa, 007; Tokuda et al., 009; Kawa, 010), whle other studes fnd a postve and sgnfcant ncome effect (Bessho and Ohkusa, 006; Babazono et al., 008; Ish, 011). The nfluence of ncome s commonly studed through estmatng the sgnfcance of the coeffcent for ncome varable, whch mght not be the most applcable approach snce t captures only lnear effects. Therefore, we follow papers that use dchotomous varables for ncome groups (Bessho and Ohkusa, 006; Tokuda et al., 009). Such approach, whch ncorporates non-lnear ncome effects, allows a comparson of the values and sgnfcance of the coeffcents for dchotomous varables of ncome groups relatve to the reference group. Snce poverty lnes vary n each Japanese muncpalty (wth muncpalty nformaton unavalable n consumer survey data), we employ ncome quntles (Ish, 011) so that the lowest quntle approxmates the low ncome group (OECD, 009). 4 The novelty of the paper s twofold. Frst, we use the 009 data for npatent and outpatent health care expendture by 4,0 adult consumers from Japan Household Panel Survey, 5 whch enables an estmaton of a mult-part model, dstngushng between non-users of health care, the users of npatent and outpatent care (Duan et al., 1983). Second, we employ a latent class approach (Deb and Trved, 1997) that better encompasses unobservable consumer characterstcs than subjectve health assessment. The mult-part model comprses equatons for the bnary choce of seekng npatent/outpatent care, as well as equatons for the amount of npatent/outpatent expendture, gven the expendture s postve. The amount of health care expendture s commonly taken n logarthms, to solve the ssues of skewness and zero mass problem (.e., the fact that health care expendture s truncated at zero). Owng to the retransformaton problem n equatons wth logged dependent varable (Duan et al., 1983; Mannng, 1998; Mullahy, 1998), lnear models can yeld unbased predctons only when error terms are normal or homoscedastc. A soluton to the retransformaton 4 Note that quntle analyss s commonly used n the studes of horzontal equty of OECD countres (van Doorslaer, 004). 4

5 problem s the use of generalzed lnear models whch specfy the mean and varance functons of the dependent varable condtonal on covarates (Nelder and Weddernburn, 197; McCullagh and Nelder, 1989). Consequently, n case of non-normalty and heteroscedastcty of error terms n OLS models for health care expendture, we use Greene s (007) generalzed lnear models wth latent classes. The results of our estmatons ndcate that the coeffcents for ncome groups are nsgnfcant both n the bnary choce models for health care use and n the models for health care expendture. Consumers separate nto two latent classes n the generalzed lnear model for outpatent health care expendture, and n the OLS model for health care expendture of those who used npatent care. In the generalzed lnear model we fnd adequate goodness-of-ft for the nverse Gaussan dstrbuton famly. As for bnary choce models, consumers do not separate nto latent classes. Overall, the results of the estmatons reveal horzontal equty of health care access and utlzaton n Japanese health nsurance system. However, horzontal nequty may be found n health nsurance premums and the prevalence of catastrophc health care coverage. The remander of the paper s structured as follows. Secton outlnes varous dmensons of equty n Japanese socal health nsurance system. Secton 3 sets up the emprcal models for measurng the demand for health care wth need and non-need varables. Secton 4 descrbes the data of Japan Household Panel Survey. The fndngs of the emprcal estmatons wth the bnary choce models, OLS and generalzed lnear models wth latent classes, along wth the analyss of the goodness-of-ft are summarzed n secton 5. Secton 6 dscusses equty n Japanese socal health nsurance system. A revew of the studes on ncome effect for health care demand n Japan, dervatons of devance resduals and Anscombe resduals, normal probablty plot for standardzed devance resduals, as well as detals on the samplng procedure n Japan Household Panel Survey are presented n the Appendx. 5 The unque feature of ths recently launched household survey s the fact that t dstngushes between npatent and outpatent health care expendture, as well as between expendture covered and non-covered 5

6 . Equty n Japanese socal health nsurance system Mandatory and unversal socal health nsurance system n Japan celebrates ts semcentennal annversary. 6 The enrolment n one of the non-ntersectng health nsurance plans s oblgatory and depends on enrollee s age and status at the labor market. The major health nsurance plans nclude: 1) natonal health nsurance, whch s muncpalty-managed nsurance for self-employed, retrees, and ther dependents; ) government-managed nsurance for small frms employees and ther dependents, and 3) company-managed nsurance assocatons formed by frms wth over 700 employees for employees and ther dependents. The year 008 saw a creaton of a specal plan for the elderly (aged 70 and above). Japanese socal health nsurance system s equtable n terms of choce of health care facltes, the sze of nomnal consurance rate, and the prces charged by provders. The users of any health nsurance plan can choose any health care nsttuton, regardless of ts locaton or type (e.g., prvate/publc, hosptal, clnc or ambulatory dvson of hosptal). There are no gatekeepers, and only n 1996 an amendment to Health Insurance Law ntroduced mnor payments for turnng to a large faclty wthout referral. 7 Whle the amount of nsurance premums s determned by each of the health nsurance plans, the types of medcal servces and drugs to be offered wthn socal health nsurance and ther costs (.e., provder prces) are set by the Mnstry of Health, Labor, and Welfare (MHLW) n a bennally revsed unfyng fee schedule. 8 The schedule ensures equal prces for smlar types of health care nsttutons. The sze of nomnal consurance rate for non-elderly and non-nfant populaton (aged 3-69) vared n the 0-50% nterval, and became a flat value of 30% for enrollees of all health nsurance plans snce 003 (Fgure 1). by socal health nsurance. 6 See the seres Japan: Unversal Health Care at 50 Years n Lancet s ssue of September 17, A payment for the frst vst to a large hosptal (wth over 00 beds) wthout referral would normally vary from 1,570 yen to 5,50 yen. 8 Wth the excepton of obstetrcs, preventve care, cosmetology and a number of addtonal types of treatment, balance bllng,.e. chargng the patent over and above the rembursement from health nsurance (Ikegam and Campbell, 004), s prohbted n Japan (Ikegam, 006). 6

7 Natonal Health Insurance Company / Government Managed Insurance 50% 50% 40% 40% 30% 30% 0% 0% 10% 10% 0% heads of household dependents 0% heads of household dependents Fgure 1. Nomnal consurance rates n Note: Nomnal consurance rate for npatent care of dependents was 0% n Although consumers pay out-of-pocket for the ncurred health care costs accordng to the nomnal consurance rate, they are compensated by ther nsurer n case of hghcost medcal expendture. The system of hgh-cost medcal benefts (catastrophc coverage) ams at enhancng ncome equty n health care access and utlzaton. Based on the amount of household ncome, consumers are compensated so that ther nomnal consurance rate become only 1% after a certan threshold value of ncurred health care expendture. As for the lowest ncome category, consumers face the cap of 35,400 yen a month, recevng the rest of the health nsurance care for free (Table 1). Owng to the system of hgh-cost medcal benefts, the values of the actual share of out-of-pocket expendture ncurred by an enrollee (effectve consurance rate) are almost twce lower than the nomnal consurance rate (Ikegam and Campbell, 1999; Ima, 00; Ikegam and Campbell, 004; Ikegam, 005). Table 1. Hgh-cost medcal benefts (catastrophc coverage) for Japanese consumers aged 3-69 Income category Hgh ncome (above 530,000 yen a month) General category Low ncome (exempt from resdence taxes) Caps on monthly out-of-pocket health nsurance expendture 150,000 yen + (health care expendture 500,000 yen)*1% <83,400 yen> 80,100 yen + (health care expendture 67,000 yen)*1% <44,400 yen> 35,400 yen <4,600 yen> Source: MHLW (011), Hese 18nen ryou sedo kakaku-kanren shryou, Kougaku ryou, kougaku kago gassan ryouyouh sedo-ntsute, Sanko shryou, p.17. Notes: Fgures n brackets correspond to the fourth hgh-cost medcal beneft wthn 1 months. All monetary values are reported accordng to the reform n October 008. The thresholds for resdence tax exemptons vary n each muncpalty. 7

8 It should be noted that the thresholds for the lowest ncome categores (those exempt from payng resdent taxes) are set at the muncpalty level. Therefore, the thresholds between the affluent (e.g., Tokyo metropoltan area) and the unprosperous muncpaltes (e.g., towns n Hokkado prefecture) may dffer up to tmes. Overall, the safety net and the thresholds are lkely to depend on the fscal stuaton n the muncpalty (Ikegam et al., 011). The studes of poverty and deprvaton n Japan have mxed results about ncome effect on the amount of health care expendture (See a revew n Appendx C). Overall, the ncome effect s rarely analyzed wth respect to ncome group. Even when ncome groups are ntroduced (e.g., Bessho and Ohkusa, 006; Tokuda et al., 009), threshold values for low and mddle-ncome groups are arbtrary chosen. 9 We beleve that employng ncome percentles (Ish, 011; OECD, 009) may be a better approach for a sample encompassng many unknown muncpaltes wth dfferent levels of poverty lnes. 3. Emprcal models Followng Gravelle et al. (006), we assume that ndvdual s welfare functon w( ) may be presented as w = w (y, x, c ), where s the ndex for consumer, y s the utlzaton of health care, x are consumer characterstcs, and c s access cost. Then, the reduced form equaton for health care utlzaton becomes y = f(x 1, x, s ), wth x 1 denotng need varables (.e., covarates that should have sgnfcant estmated coeffcents), 10 x standng for non-need varables (.e., covarates that should not have an effect on health care utlzaton), 11 and s ndcatng supply varables (e.g., per capta number of doctors or number of beds). Below we outlne econometrc models for health care access (the bnary choce of gong to clnc/hosptal) and utlzaton (the amount of health care expendture). To address ncome equty n health care access and ncurred health care costs (Culyer and Wagstaff, 1993), our emprcal analyss focuses on the examnaton of the estmated coeffcents for ncome groups. 9 Bessho and Ohkusa (006) separate consumers nto the followng ncome groups wth respect to annual household ncome: les than 1 mln. Yen, 1-3 mln. Yen, 3-5 mln. Yen, 5-7 mln. Yen, 7-10 mln. Yen, mln. Yen, 15-0 mln. Yen, 0-30 mln. Yen, above 30 mln. Yen. At the same tme, Tokuda et al. (009) create categores: less than 1.48 mln. Yen, mln. Yen, 4-6 mln. Yen, above 6 mln. Yen. Note that the annual CPI nflaton n the perod between the two studes less than 1%. 10 E.g., morbdty or self-assessed health. 11 E.g., ncome. 8

9 3.1 Mult-part models The four-part model dstngushes between non-users of health care, users of npatent and outpatent care. The model ncorporates bnary choce equatons and s estmated usng maxmum lkelhood method, wth each equaton of the model estmated separately owng to an addtve log-lkelhood functon (Duan et al., 1983). Let Pr( y >0 ) = F( y, x ' β 1 ) (1) Pr( npatent > 0 y >0 ) = F( npatent, x β ' ) () log (y y >0, npatent =0) = x ' β 3 + ε (3) log (y npatent >0) = x β ' 4 + ν (4) Eε =Eν = E (x ε ' )= E (x ν ' )=0, (5) where s the ndex for observatons, y denotes health care expendture, npatent ndcates npatent health care expendture, and x are covarates. The dependent varables n (3) and (4) are taken n logs due to the skewness of health expendture data and zero mass problem (.e., the fact that health care expendture s truncated at zero). The four-part model (1)-(5) s an extenson of the (1), (3), (5) two-part model (Duan et al., 1983; Duan et al., 1984) as specfed below: Pr(y >0)=F(y, x γ ' 1 ) (1) log (y ) = x ' γ + ξ (3) Eξ = E (x ξ ' )= 0, (5) where s the ndex for observatons, y denotes health care expendture, and x are covarates. 3. Generalzed lnear models Owng to the retransformaton problem n regressons wth logged dependent varable (Duan, 1983; Mannng, 1998; Mullahy, 1998), estmatng lnear models (3) and (4) can yeld unbased predctons only when error terms are normal or homoscedastc. More formally, n terms of notatons for equaton (3), f ε ~ N(0, σ ε I), then E(y x) = exp(x ' β σ I). If ε ε are not normal, but..d., then E(y x) = exp(x ' β 3 ) E(exp(ε)), and therefore, ^ E (y x)= exp(x ' 3 ) E(exp( )). However, the estmate of E(y x) becomes based n case of heteroscedastc errors. Indeed, when varance s some functon v( ) of 9

10 covarates, namely Var(ε)=v(x), the expresson for the expectancy of y condtonal on x becomes E(y x)=exp(x ' β 3 ) v(x). A soluton to the retransformaton problem n case of non-normal and heteroskedastc errors s the use of generalzed lnear models (Nelder and Wedderburn, 197; McCullagh and Nelder, 1989) for health care expendture data (Mullahy, 1998; Blough et al., 1999). Although there are other possble solutons, 1 the advantages of generalzed lnear models are mproved precson compared to OLS-methods and robustness of the estmate of the condtonal mean (Mannng and Mullahy, 001). Generalzed lnear models assume a partcular form of dstrbuton famly, whch requres postestmaton analyss about the goodness-of-ft. Generalzed lnear model specfes the mean and varance functons for y x by settng a famly of dstrbutons g( ), as well as the lnk functon f( ), so that f(e(y x)) = x ' β. We use LIMDEP 9.0 to analyze the models for nonnegatve dependent varables wth lognormal, gamma, Webull, and nverse Gaussan famles. Let f(e(y x))= x ' β (6) y x ~ g(y, x ' β, θ), (7) where f( ) denotes a logarthmc lnk functon, g( ) s a famly of dstrbuton, x are covarates, and θ are ancllary parameters. For each dstrbuton famly we examne the model ft, employng normalty test of Anscombe resduals (McCullagh and Nelder, 1989; Dobson, 00; Agrest, 007) and standardzed devance resduals (Davson and Ggl, 1989). 13 The comparson of the goodness-of-ft between OLS and generalzed lnear models s conducted wth the analyss of resduals (raw bas and mean squared error). 1 There are several alternatve ways to deal wth heteroscedastcty. Among them are Mannng s (1998) method, whch s partcularly easy to mplement f heteroscedastcty s present across mutually exclusve groups; sem-parametrc approaches and extenstons of generalzed lnear models (Basu and Mannng, 009). Recent revews of the appled lterature wth generalzed lnear models and other methods for modelng health care expendture may be found n Mhaylova et al. (011), Mullahy (009), Basu and Mullahy (009), Buntn and Zaslavsky (004). 13 See dervaton of model devance and devance resduals n the Appendx. 10

11 3.3 Latent class analyss Bnary choce model wth latent classes The latent class approach (Deb and Trved, 1997; Deb and Holmes, 000) dvdes consumers nto unobservable classes of hgh and low users of health care to account for mmeasurable consumer characterstcs, not captured by self-assessed health and other varables. The bnary choce model (1) s extended to a latent class model n the followng way: Pr(y >0)=F(y, x β ' 1j ), (8) where s the ndex for observatons, j s the ndex for latent class (j =1 J), y s health care expendture, x are covarates related to the demand for health care, β 1j are coeffcents for j-th latent class. The estmatons are conducted n LIMDEP 9.0, whch determnes the most probable latent class by comparng posteror jont probabltes Pr(j ) for all j-s, wth the pror probablty F j of belongng to latent class j and posteror jont probablty Pr(j ) of belongng to latent class j calculated as: F j = J 1 j = 1 expϑ (1 + expϑ ) j j (9) Pr(j )= F Pr( j) J j= 1 j F Pr( j) j, (10) where Pr( j) s the densty functon of y gven observaton belongs to class j. Equatons ()-(4) are transformed nto a latent class model n a smlar way Generalzed lnear models wth latent classes For generalzed lnear models that ft the data, equatons (6)-(7) are extended as follows: f(e(y x))= x ' β j (11) y x ~ g(y, x ' β j, θ j ), (1) where f( ) denotes a logarthmc lnk functon, g( ) s a famly of dstrbuton, x are covarates, j s the ndex for latent class (j = 1 J), y s health care expendture, β j are coeffcents, θ j are ancllary parameters. The pror and posteror class probabltes are calculated accordng to (9) and (10). 11

12 3.3.3 Specfcaton tests Greene (007) proposes the followng statstcs to test between H o : a latent class (unrestrcted) model and H a : a model wthout latent classes (restrcted model) : L = (lnl u -lnl R ) ~ χ ( -1) (k 1) ) (J +, (13) where lnl u s loglkelhood of the unrestrcted model, lnl R s loglkelhood of the restrcted model, J s the number of latent classes, and k s the number of covarates. Although the statstcs L corresponds to the general logcs of lkelhood rato test for nested models, Greene (007) argues that the valdty of the statstcs needs to be further nvestgated, and the use of conventonal nformaton crtera s more preferable n the appled analyss. 14 Therefore, to choose between the models wth and wthout latent classes, we use both Greene s (007) LR test as specfed n (13) and nformaton crtera (AIC and BIC). 4. Data The paper uses the data of Japan Household Panel Survey. The survey was establshed n 009 as a natonally representatve annual survey of adults. Respondents aged above 0 answer a wde range of questons on ther labor actvty, ncome and expendture, soco-demographc characterstcs, anthropometry, health, and health-related lfestyles. There are a number of unque features of ths longtudnal survey for the purposes of the analyss of health care demand. Frst, health care utlzaton s reported at the ndvdual level. 15 Second, health care utlzaton s dvded nto health care n outpatent and npatent facltes. Fnally, health care expendture s subdvded nto the expendture covered and uncovered by health nsurance. The partcpaton n our analyss s modeled through dchotomous varables healthcare for usng any health care faclty (corresponds to eq.1 n 3.1), and npatent care for seekng care n an npatent faclty gven consumer used some health care faclty (eq.3 n 3.1). The ntensty varable expendture s out-of-pocket payments for health care covered by health nsurance (eq. 3 and 4 n 3.1). We construct dchotomous varables group 1 through group 5 for quntles of the annual dsposable (after-tax) household ncome (wth the upper quntle group 5 treated as a reference category). Fve nteracton terms (ncome group*log of annual 14 Greene (007) Testng for the Latent Class Model. In: LIMDEP. Verson 9.0. Econometrc modelng gude. Vol.1. E Whle n Japanese Panel Survey of Consumers and Keo Household Panel Survey health care expendture s reported at the household level. 1

13 dsposable ncome) are added to the lst of regressors to estmate ncome elastcty n each quntle. Indvdual characterstcs are age, gender, bnary varables for graduate educaton, and employment. Health status s taken nto account wth a bnary varable for low health condton, Ben-Sra s (198) psychologcal dstress ndex (PDI), and body mass ndex (BMI). Bnary varables for drnkng, smokng, sports, and checkups reflect health-related lfe styles. The bnary varables for desgnated cty and other ctes capture health care supply whch s generally better n Japanese urban areas (rural areas,.e., towns and vllages become a reference category). We add a dummy for Natonal Health Insurance, snce sometmes there are addtonal hgh-cost medcal benefts for the poor n ths health nsurance plan. We use a subsample of non-elderly consumers (aged below 70), snce Japanese elderly have lower nomnal consurance rates 16 and specal thresholds for hgh-cost medcal benefts (Table ). 16 Snce 007 nomnal consurance rate s 10% for aged above 75 and 0% for aged

14 Table. Descrptve statstcs of our sample Varable Defnton Obs Mean St.Dev. Mn Max healthcare = 1 f out-of-pocket expendture for health care covered by health nsurance s nonnegatve n 008; 0 otherwse npatent care expendture = 1 f out-of-pocket expendture for npatent care covered by health nsurance s nonnegatve n 008 gven ntensty equals 1; 0 otherwse out-of-pocket expendture for health care covered by health nsurance n 008, thousand yen ncome dsposable household ncome n 008, thousand yen age years of age as of January 31, gender =1 f female; 0 f male educaton = 1 f completed junor college, college or unversty work = 1 f was employed last month desgnated cty = 1 f lves n a desgnated cty, 0 otherwse cty = 1 f lve n a non-desgnated cty, 0 otherwse lowhcond PDI BMI =1 f self-assessed health condton s reported as not very healthy or not at all healthy ; 0 f self-assessed health condton s reported as very healthy, rather healthy or average health physologcal dstress ndex, calculated as the sum of responses to the questons on the recent presence of the below twelve condtons (each response s gven on a four-pont scale, where one refers to often, two means sometmes, three mples almost never, and four stands for never ): headache or dzzness; palptaton or shortness of breath; senstve stomach and ntestnes; backache or shoulder pan; get tred easly; catch a cold easly often feel rrtated; trouble gettng to sleep; feel reluctant to meet people; less concentraton on work; dssatsfed wth present lfe; anxety over the future weght( kg) body mass ndex = heght ( cm ) smokng = 1 f currently smokes; 0 otherwse drnkng = 1 f drnks moderately or heavly; 0 otherwse NHI = 1 f Natonal Health Insurance; 0 otherwse (other health nsurance plan) checkup gym = 1 f had nonnegatve expendture for varous checkups n 008 (apart from checkups at work); 0 otherwse = 1 f had nonnegatve expendture for dong sports, gong to gym, and buyng supplements n 008; 0 otherwse

15 5. Emprcal results 5.1 Bnary choce model for health care utlzaton Accordng to the results of the test for normalty of errors (Greene, 007), 17 we use probt model for bnary choce equatons of the four-part model (eq.1 and eq.). For each equaton we estmate a model wth two latent classes. The pror probabltes for latent class membershp are sgnfcant and Greene s (007) lkelhood rato test rejects the null hypothess of the model wthout latent classes. Yet, n each case we could not conclude that consumers separate nto two latent classes wth respect to ther bnary choce of seekng health care. 18 Indeed, AIC and BIC for the models wth and wthout latent classes are close. Moreover, margnal effects for most of explanatory varables n each latent class are nsgnfcant. Consequently, for each equaton we estmate probt model wthout latent classes (Table 3). The results reveal that wth the excepton of the forth ncome quntle n eq.1, the coeffcents for margnal effects for ncome groups are nsgnfcant n both eq.1 and eq.. Moreover, most of other non-need varables are nsgnfcant. Age s the only sgnfcant determnant of the bnary choce for seekng npatent care. In case of any type of health care, the sgnfcant covarates are age, gender, graduate educaton, and some lfestyle varables: body mass ndex and the bnary varable for checkups. 17 Greene (007). A Test for Normalty n the Probt Model In: LIMDEP 9.0. Econometrc modelng gude. Vol.1. E The result s smlar to the prevous fndng wth the data on Japanese women, where consumers dd not separate nto latent classes n the bnary choce model for seekng health care (Besstremyannaya, 011). 15

16 Table 3. Margnal effects n the bnary choce equatons (1) and () of a four-part model (1) () Healthcare Inpatent care constant (0.5306) (0.046)* age (0.0006) *** (0.0003)*** group 1 * ln(ncome) (0.0356) (0.0136) group * ln(ncome) (0.1570) (0.0664) group 3 * ln(ncome) (0.840) (0.1187) group 4 * ln(ncome) (0.03) *** (0.0859) group 5 * ln(ncome) (0.0579) (0.03) PDI ( ) ( ) BMI ( )*** (0.0000) gender (0.0169) *** (0.0071) educaton (0.0177) *** (0.0076) lowhcond (0.000) (0.0001) smokng (0.0001) ( ) drnkng (0.0001) ( ) NHI (0.0193) (0.0079) checkup (0.0001) *** (0.0000) gym ( ) (0.0000) work (0.000) (0.0001) desgnated cty (0.0316) (0.0100) cty (0.083) (0.01) group (0.588) (0.48) group (1.665) (0.5395) group (.103) (0.8946) group (1.6576) *** (0.718) Log lkelhood Observatons Notes: *** p< 0.01, ** p< 0.05, *p< 0.1. Robust standard errors n parentheses. Margnal effects are evaluated at sample means. Group 1, group, group 3, group 4 and group 5 denote dchotomous varables for log(ncome) quntles, wth group 1 standng for the lowest quntle, and group 5 ndcatng the hghest quntle. 5. Modelng health care expendture wth logged dependent varable The post-estmaton analyss wth an ordnary least squares model for equaton (3) reveals that the errors are non-normal and heteroscedastc. Consequently, we experment wth generalzed lnear models wth four dstrbuton famles: lognormal, gamma, Webull, and nverse Gaussan. The results of the resdual analyss ndcate that nverse Gaussan dstrbuton provdes the best model ft n terms of the bas, mean squared error, and Anscombe resduals (Table 4, Fg.-3). 16

17 Table 4. Model comparson Lnear Model Generalzed lnear models (a) (b) lognormal (c) gamma (d) Webull dstrbuton dstrbuton dstrbuton (e) nverse Gaussan dstrbuton Mean raw bas (resdual) Mean squared error Normalty test, Anscombe resduals Notes: In lnear model the ftted values are calculated wth the smearng factor. Snce the general form of Webull famly does not lead to convergence, we use Raylegh dstrbuton (.e., the scale parameter n Webull dstrbuton equals two). Normalty test reports the p-value for jont probablty n skewness/kurtoss test wth the null hypothess of the standard normal dstrbuton. As for standardzed devance resduals, standardzed resduals, and Person resduals, the null hypothess of normalty s not accepted n all the generalzed lnear models. Dchotomous varables for ncome groups are excluded from the lst of covarates n generalzed lnear models snce they nfluenced convergence (namely, the margnal effects for these varables were huge). Although the dstrbuton of standardzed devance resduals s close to normal (See Fg.4 n Appendx A), the skewness/kurtoss test rejected the null hypothess of normalty Resduals verses ftted values RESIDUAL YF IT T E D Fgure. Resduals verses ftted values for the generalzed lnear model wth nverse Gaussan dstrbuton 17

18 7.1 7 N ormal probablty plot for A nscombe resduals ANSCOMBE Qu a n tle N-Q_ p lo t o f ANSCOM BEv s. N( , ) Fgure 3. Quntles of Anscombe resduals verses quntles of normal dstrbuton for the generalzed lnear model wth nverse Gaussan dstrbuton 5.3 Income equty n a model wth latent classes Consumers who used only outpatent care We estmate a generalzed lnear model wth nverse Gaussan dstrbuton and two latent classes, and fnd that the coeffcents for latent class probabltes are sgnfcant (Table 5). Accordng to the results of Greene s (007) LR test for nested models, H o of unrestrcted model (wth latent classes) s not rejected. Smlarly, the comparson of nformaton crtera demonstrates that the model wth latent classes s preferred to the model wthout latent classes. Consequently, we may conclude that consumers separate nto two latent classes wth respect to ther outpatent expendture. The frst latent class (183 observatons) s relatvely young adults: mean age 44.06, standard devaton Only 5% of them have low health condton. The average annual outpatent health care expendture of the frst latent class s 61,377 yen, however, the standard devaton of ths varable s hgh: 197,163 yen. The second latent class contans 857 observatons for relatvely older adults: mean age 50.98, standard devaton The prevalence of low health condton n the second latent class s 15.5%. The average annual outpatent health care expendture of the second latent class s 60,435 yen, whch s close to the value of ths varable n the frst latent class. 18

19 However, the standard devaton of the varable n the second latent class s tmes smaller than n the frst latent class: 80,301 yen. The coeffcents for ncome groups are nsgnfcant n each of the latent classes. Ths mples that smlarly to our fndngs for the bnary choce models, there s horzontal equty n ncome n each of the latent classes. The need varables (age and low health condton) are sgnfcant covarates n each latent class. The fndngs on horzontal equty n Japan may be contrasted to the estmatons of log health care expendture of the US elderly n a lnear model wth two latent classes, where the coeffcents for the lowest ncome quartle are sgnfcant n each class (Deb and Trved, 011). Table 5. Estmatng health care expendture wth a generalzed lnear model wth nverse Gaussan dstrbuton and latent classes (consumers who used only outpatent care) Whole sample Latent class 1 Latent class constant (0.689)*** (.3896) (0.8433) age (0.0037)*** (0.0083)*** (0.0040)*** group 1 * ln(ncome) (0.0157) (0.880) (0.1057) group * ln(ncome) (0.0095) (0.53) (0.0959) group 3 * ln(ncome) (0.0143) (0.50) (0.0935) group 4 * ln(ncome) (0.0118) (0.381) (0.0900) group 5 * ln(ncome) (0.0103)** (0.39) (0.0858) PDI (0.0003) (0.010)* (0.0060) BMI (0.0005) (0.09) (0.0113) gender (0.093) (0.168) 0.95 (0.0937)** educaton (0.1045)* (0.189) (0.0938) lowhcond (0.085) (0.5503)*** (0.1374)*** smokng (0.0115) (0.579) (0.095)* drnkng (0.0009) (0.373) (0.088) NHI (0.140) (0.95) (0.0953) checkup (0.0006) (0.93)*** (0.1093)*** gym (0.0005) (0.198) (0.0877)*** work (0.011) (0.66) (0.1140)*** desgnated cty (0.1876) (0.4855) (0.1646) cty (0.1768) (0.4660)** (0.1595) Log lkelhood Observatons Scale parameter n the dstrbuton (0.1978)*** (0.395)*** Pror probablty for class membershp (0.0359)*** (0.0359)*** Notes: The dependent varable s annual health care expendture. The Table reports coeffcents for covarates n condtonal mean functon, and robust standard errors n parentheses. *** p< 0.01, ** p< 0.05, *p<

20 5.3. Consumers who used npatent care The results of the heteroscedastcty test ndcate that the errors n the ordnary least squares models for health care expendture of consumers who used npatent care (eq.4) are homoscedastc. Consequently, we do not use generalzed lner models and employ an OLS model wth latent classes. Snce the subsample of npatent care users s 141 consumers, we keep the mnmal number of covarates. Namely, the regressors are age, gender, the bnary varable for low health condton, and the dummes for ncome quntles. The results of the estmatons reveal nsgnfcance of ncome groups n each latent class (Table 6). In other words, horzontal equty s found for health care expendture of Japanese consumers who used npatent care. Table 6. Estmatng a latent class lnear model for consumers who used npatent health care Whole sample Latent class 1 Latent class constant ( )*** ( )*** (5.67) age (1.8454) ( )*** (0.8366)** gender (5.1148) ( ) (1.534) group ( ) ( ) (5.9890) group (5.159) ( ) (3.4999) group ( ) ( ) ( ) group ( ) (608.97) (6.9913) lowhcond (0.384) ( ) (4.9714)* Log lkelhood Observatons Pror probablty for class membershp (0.0439)*** (0.0439)*** Note: The dependent varable s logarthm of annual health care expendture for the subsample that used npatent care. The Table reports coeffcents for covarates and robust standard errors n parentheses. *** p< 0.01, ** p< 0.05, *p< Dscusson Our estmatons, whch account for unobservable consumer heterogenety through a latent class approach, reveal horzontal equty n health care access and the amount of out-of-pocket expendture for health care covered by Japanese socal health nsurance. Overall, the presence of horzontal equty n health care access and utlzaton n Japan s smlar to the fndngs on equtable or pro-poor non-specalst care utlzaton n OECD countres (van Doorslaer et al., 004). Moreover, n terms of total health care expendture of consumers, socal health nsurance system n Japan s found to be more equtable than n Germany (Ikegam et al., 011). However, there are other aspects where Japanese socal health nsurance system demonstrates ncome nequty: health nsurance premums and catastrophc coverage (Ikegam et al., 011; Hashmoto et al., 011; HGPI, 009). 0

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

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

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

On Geographic Inequality in Japanese Regional Health Insurance

On Geographic Inequality in Japanese Regional Health Insurance On Geographc Inequalty n Japanese Regonal Health Insurance Narmasa KUMAGAI Faculty of Economcs Knk Unversty 10 February 2010 Abstract In Japan, economc stagnaton due to the lack of aggregate demand has

More information

Binomial Link Functions. Lori Murray, Phil Munz

Binomial Link Functions. Lori Murray, Phil Munz Bnomal Lnk Functons Lor Murray, Phl Munz Bnomal Lnk Functons Logt Lnk functon: ( p) p ln 1 p Probt Lnk functon: ( p) 1 ( p) Complentary Log Log functon: ( p) ln( ln(1 p)) Motvatng Example A researcher

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

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

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

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

THE DISTRIBUTION OF LOAN PORTFOLIO VALUE * Oldrich Alfons Vasicek

THE DISTRIBUTION OF LOAN PORTFOLIO VALUE * Oldrich Alfons Vasicek HE DISRIBUION OF LOAN PORFOLIO VALUE * Oldrch Alfons Vascek he amount of captal necessary to support a portfolo of debt securtes depends on the probablty dstrbuton of the portfolo loss. Consder a portfolo

More information

Financial Instability and Life Insurance Demand + Mahito Okura *

Financial Instability and Life Insurance Demand + Mahito Okura * Fnancal Instablty and Lfe Insurance Demand + Mahto Okura * Norhro Kasuga ** Abstract Ths paper estmates prvate lfe nsurance and Kampo demand functons usng household-level data provded by the Postal Servces

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

! # %& ( ) +,../ 0 1 2 3 4 0 4 # 5##&.6 7% 8 # 0 4 2 #...

! # %& ( ) +,../ 0 1 2 3 4 0 4 # 5##&.6 7% 8 # 0 4 2 #... ! # %& ( ) +,../ 0 1 2 3 4 0 4 # 5##&.6 7% 8 # 0 4 2 #... 9 Sheffeld Economc Research Paper Seres SERP Number: 2011010 ISSN 1749-8368 Sarah Brown, Aurora Ortz-Núñez and Karl Taylor Educatonal loans and

More information

Traffic-light a stress test for life insurance provisions

Traffic-light a stress test for life insurance provisions MEMORANDUM Date 006-09-7 Authors Bengt von Bahr, Göran Ronge Traffc-lght a stress test for lfe nsurance provsons Fnansnspetonen P.O. Box 6750 SE-113 85 Stocholm [Sveavägen 167] Tel +46 8 787 80 00 Fax

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

The Choice of Direct Dealing or Electronic Brokerage in Foreign Exchange Trading

The Choice of Direct Dealing or Electronic Brokerage in Foreign Exchange Trading The Choce of Drect Dealng or Electronc Brokerage n Foregn Exchange Tradng Mchael Melvn & Ln Wen Arzona State Unversty Introducton Electronc Brokerage n Foregn Exchange Start from a base of zero n 1992

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

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

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

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

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

How To Find Out If A Span Or Spanards Is Healthier

How To Find Out If A Span Or Spanards Is Healthier HEDG Workng Paper 08/22 A comparson of the health status and health care utlsaton patterns between foregners and the natonal populaton n Span: new evdence from the Spansh Natonal Health Survey C Hernández

More information

Analysis of Premium Liabilities for Australian Lines of Business

Analysis of Premium Liabilities for Australian Lines of Business Summary of Analyss of Premum Labltes for Australan Lnes of Busness Emly Tao Honours Research Paper, The Unversty of Melbourne Emly Tao Acknowledgements I am grateful to the Australan Prudental Regulaton

More information

Cahiers de la Chaire Santé

Cahiers de la Chaire Santé Cahers de la Chare Santé The nfluence of supplementary health nsurance on swtchng behavour: evdence from Swss data Auteurs : Brgtte Dormont, Perre-Yves Geoffard, Karne Lamraud N 4 - Janver 2010 1 The nfluence

More information

Health Insurance and Household Savings

Health Insurance and Household Savings Health Insurance and Household Savngs Mnchung Hsu Job Market Paper Last Updated: November, 2006 Abstract Recent emprcal studes have documented a puzzlng pattern of household savngs n the U.S.: households

More information

Scale Dependence of Overconfidence in Stock Market Volatility Forecasts

Scale Dependence of Overconfidence in Stock Market Volatility Forecasts Scale Dependence of Overconfdence n Stoc Maret Volatlty Forecasts Marus Glaser, Thomas Langer, Jens Reynders, Martn Weber* June 7, 007 Abstract In ths study, we analyze whether volatlty forecasts (judgmental

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

PUBLIC VS. PRIVATE HEALTH CARE SERVICES DEMAND IN ITALY. A COUNT DATA ANALYSIS ON SHAW DATA

PUBLIC VS. PRIVATE HEALTH CARE SERVICES DEMAND IN ITALY. A COUNT DATA ANALYSIS ON SHAW DATA XIV CONFERENZA IL FUTURO DEI SISTEMI DI WELFARE NAZIONALI TRA INTEGRAZIONE EUROPEA E DECENTRAMENTO REGIONALE coordnamento, competzone, mobltà Pava, Unverstà, 4-5 ottobre 2002 PUBLIC VS. PRIVATE HEALTH

More information

Evaluating the Effects of FUNDEF on Wages and Test Scores in Brazil *

Evaluating the Effects of FUNDEF on Wages and Test Scores in Brazil * Evaluatng the Effects of FUNDEF on Wages and Test Scores n Brazl * Naérco Menezes-Flho Elane Pazello Unversty of São Paulo Abstract In ths paper we nvestgate the effects of the 1998 reform n the fundng

More information

Searching and Switching: Empirical estimates of consumer behaviour in regulated markets

Searching and Switching: Empirical estimates of consumer behaviour in regulated markets Searchng and Swtchng: Emprcal estmates of consumer behavour n regulated markets Catherne Waddams Prce Centre for Competton Polcy, Unversty of East Angla Catherne Webster Centre for Competton Polcy, Unversty

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

Willingness to Pay for Health Insurance: An Analysis of the Potential Market for New Low Cost Health Insurance Products in Namibia

Willingness to Pay for Health Insurance: An Analysis of the Potential Market for New Low Cost Health Insurance Products in Namibia Wllngness to Pay for Health Insurance: An Analyss of the Potental Market for New Low Cost Health Insurance Products n Namba Abay Asfaw Center for Dsease Control and Preventon\Natonal Insttute for Occupatonal

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

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

Start me up: The Effectiveness of a Self-Employment Programme for Needy Unemployed People in Germany*

Start me up: The Effectiveness of a Self-Employment Programme for Needy Unemployed People in Germany* Start me up: The Effectveness of a Self-Employment Programme for Needy Unemployed People n Germany* Joachm Wolff Anton Nvorozhkn Date: 22/10/2008 Abstract In recent years actvaton of means-tested unemployment

More information

High Correlation between Net Promoter Score and the Development of Consumers' Willingness to Pay (Empirical Evidence from European Mobile Markets)

High Correlation between Net Promoter Score and the Development of Consumers' Willingness to Pay (Empirical Evidence from European Mobile Markets) Hgh Correlaton between et Promoter Score and the Development of Consumers' Wllngness to Pay (Emprcal Evdence from European Moble Marets Ths paper shows that the correlaton between the et Promoter Score

More information

The demand for private health care in the UK

The demand for private health care in the UK Journal of Health Economcs 19 2000 855 876 www.elsever.nlrlocatereconbase The demand for prvate health care n the UK Carol Propper ) Department of Economcs, CASE and CEPR, UnÕersty of Brstol, Brstol BS8

More information

ADVERSE SELECTION IN INSURANCE MARKETS: POLICYHOLDER EVIDENCE FROM THE U.K. ANNUITY MARKET *

ADVERSE SELECTION IN INSURANCE MARKETS: POLICYHOLDER EVIDENCE FROM THE U.K. ANNUITY MARKET * ADVERSE SELECTION IN INSURANCE MARKETS: POLICYHOLDER EVIDENCE FROM THE U.K. ANNUITY MARKET * Amy Fnkelsten Harvard Unversty and NBER James Poterba MIT and NBER * We are grateful to Jeffrey Brown, Perre-Andre

More information

Returns to Experience in Mozambique: A Nonparametric Regression Approach

Returns to Experience in Mozambique: A Nonparametric Regression Approach Returns to Experence n Mozambque: A Nonparametrc Regresson Approach Joel Muzma Conference Paper nº 27 Conferênca Inaugural do IESE Desafos para a nvestgação socal e económca em Moçambque 19 de Setembro

More information

------ Evidence from a Natural Experiment in China

------ Evidence from a Natural Experiment in China Are Assets n Medcal Savngs Accounts Dscounted? ------ Evdence from a Natural Experment n Chna Maoyong Fan, Zhen Le and Guoen Lu November 23rd, 2010 Abstract: In Chna, Medcal Savngs Accounts (MSAs) are

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

To manage leave, meeting institutional requirements and treating individual staff members fairly and consistently.

To manage leave, meeting institutional requirements and treating individual staff members fairly and consistently. Corporate Polces & Procedures Human Resources - Document CPP216 Leave Management Frst Produced: Current Verson: Past Revsons: Revew Cycle: Apples From: 09/09/09 26/10/12 09/09/09 3 years Immedately Authorsaton:

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

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

TESTING FOR EVIDENCE OF ADVERSE SELECTION IN DEVELOPING AUTOMOBILE INSURANCE MARKET. Oksana Lyashuk

TESTING FOR EVIDENCE OF ADVERSE SELECTION IN DEVELOPING AUTOMOBILE INSURANCE MARKET. Oksana Lyashuk TESTING FOR EVIDENCE OF ADVERSE SELECTION IN DEVELOPING AUTOMOBILE INSURANCE MARKET by Oksana Lyashuk A thess submtted n partal fulfllment of the requrements for the degree of Master of Arts n Economcs

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

Stress test for measuring insurance risks in non-life insurance

Stress test for measuring insurance risks in non-life insurance PROMEMORIA Datum June 01 Fnansnspektonen Författare Bengt von Bahr, Younes Elonq and Erk Elvers Stress test for measurng nsurance rsks n non-lfe nsurance Summary Ths memo descrbes stress testng of nsurance

More information

DO LOSS FIRMS MANAGE EARNINGS AROUND SEASONED EQUITY OFFERINGS?

DO LOSS FIRMS MANAGE EARNINGS AROUND SEASONED EQUITY OFFERINGS? DO LOSS FIRMS MANAGE EARNINGS AROUND SEASONED EQUITY OFFERINGS? Fernando Comran, Unversty of San Francsco, School of Management, 2130 Fulton Street, CA 94117, Unted States, fcomran@usfca.edu Tatana Fedyk,

More information

WORKING PAPERS. The Impact of Technological Change and Lifestyles on the Energy Demand of Households

WORKING PAPERS. The Impact of Technological Change and Lifestyles on the Energy Demand of Households ÖSTERREICHISCHES INSTITUT FÜR WIRTSCHAFTSFORSCHUNG WORKING PAPERS The Impact of Technologcal Change and Lfestyles on the Energy Demand of Households A Combnaton of Aggregate and Indvdual Household Analyss

More information

Evaluating the generalizability of an RCT using electronic health records data

Evaluating the generalizability of an RCT using electronic health records data Evaluatng the generalzablty of an RCT usng electronc health records data 3 nterestng questons Is our RCT representatve? How can we generalze RCT results? Can we use EHR* data as a control group? *) Electronc

More information

APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT

APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT Toshhko Oda (1), Kochro Iwaoka (2) (1), (2) Infrastructure Systems Busness Unt, Panasonc System Networks Co., Ltd. Saedo-cho

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 Choice of Direct Dealing or Electronic Brokerage in Foreign Exchange Trading

The Choice of Direct Dealing or Electronic Brokerage in Foreign Exchange Trading The Choce of Drect Dealng or Electronc Brokerage n Foregn Exchange Tradng Mchael Melvn Arzona State Unversty & Ln Wen Unversty of Redlands MARKET PARTICIPANTS: Customers End-users Multnatonal frms Central

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

Does Higher Education Enhance Migration?

Does Higher Education Enhance Migration? DISCUSSION PAPER SERIES IZA DP No. 7754 Does Hgher Educaton Enhance Mgraton? Mka Haapanen Petr Böckerman November 2013 Forschungsnsttut zur Zukunft der Arbet Insttute for the Study of Labor Does Hgher

More information

How To Study The Nfluence Of Health Insurance On Swtchng

How To Study The Nfluence Of Health Insurance On Swtchng Workng Paper n 07-02 The nfluence of supplementary health nsurance on swtchng behavour: evdence on Swss data Brgtte Dormont, Perre- Yves Geoffard, Karne Lamraud The nfluence of supplementary health nsurance

More information

Covariate-based pricing of automobile insurance

Covariate-based pricing of automobile insurance Insurance Markets and Companes: Analyses and Actuaral Computatons, Volume 1, Issue 2, 2010 José Antono Ordaz (Span), María del Carmen Melgar (Span) Covarate-based prcng of automoble nsurance Abstract Ths

More information

Estimation of Dispersion Parameters in GLMs with and without Random Effects

Estimation of Dispersion Parameters in GLMs with and without Random Effects Mathematcal Statstcs Stockholm Unversty Estmaton of Dsperson Parameters n GLMs wth and wthout Random Effects Meng Ruoyan Examensarbete 2004:5 Postal address: Mathematcal Statstcs Dept. of Mathematcs Stockholm

More information

HARVARD John M. Olin Center for Law, Economics, and Business

HARVARD John M. Olin Center for Law, Economics, and Business HARVARD John M. Oln Center for Law, Economcs, and Busness ISSN 1045-6333 ASYMMETRIC INFORMATION AND LEARNING IN THE AUTOMOBILE INSURANCE MARKET Alma Cohen Dscusson Paper No. 371 6/2002 Harvard Law School

More information

On the Optimal Control of a Cascade of Hydro-Electric Power Stations

On the Optimal Control of a Cascade of Hydro-Electric Power Stations On the Optmal Control of a Cascade of Hydro-Electrc Power Statons M.C.M. Guedes a, A.F. Rbero a, G.V. Smrnov b and S. Vlela c a Department of Mathematcs, School of Scences, Unversty of Porto, Portugal;

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

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

How To Understand The Results Of The German Meris Cloud And Water Vapour Product

How To Understand The Results Of The German Meris Cloud And Water Vapour Product Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller

More information

1.1 The University may award Higher Doctorate degrees as specified from time-to-time in UPR AS11 1.

1.1 The University may award Higher Doctorate degrees as specified from time-to-time in UPR AS11 1. HIGHER DOCTORATE DEGREES SUMMARY OF PRINCIPAL CHANGES General changes None Secton 3.2 Refer to text (Amendments to verson 03.0, UPR AS02 are shown n talcs.) 1 INTRODUCTION 1.1 The Unversty may award Hgher

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

Hollinger Canadian Publishing Holdings Co. ( HCPH ) proceeding under the Companies Creditors Arrangement Act ( CCAA )

Hollinger Canadian Publishing Holdings Co. ( HCPH ) proceeding under the Companies Creditors Arrangement Act ( CCAA ) February 17, 2011 Andrew J. Hatnay ahatnay@kmlaw.ca Dear Sr/Madam: Re: Re: Hollnger Canadan Publshng Holdngs Co. ( HCPH ) proceedng under the Companes Credtors Arrangement Act ( CCAA ) Update on CCAA Proceedngs

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

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

The impact of hard discount control mechanism on the discount volatility of UK closed-end funds

The impact of hard discount control mechanism on the discount volatility of UK closed-end funds Investment Management and Fnancal Innovatons, Volume 10, Issue 3, 2013 Ahmed F. Salhn (Egypt) The mpact of hard dscount control mechansm on the dscount volatlty of UK closed-end funds Abstract The mpact

More information

A DYNAMIC ANALYSIS OF

A DYNAMIC ANALYSIS OF A DYNAMIC ANALYSIS OF THE DEMAND FOR LIFE INSURANCE Andre P. Lebenberg (contact author) Faculty of Fnance The Unversty of Msssspp Oxford, MS 38677 alebenberg@bus.olemss.edu Tel: 662.915.3844 James M. Carson

More information

L10: Linear discriminants analysis

L10: Linear discriminants analysis L0: Lnear dscrmnants analyss Lnear dscrmnant analyss, two classes Lnear dscrmnant analyss, C classes LDA vs. PCA Lmtatons of LDA Varants of LDA Other dmensonalty reducton methods CSCE 666 Pattern Analyss

More information

The Economic Impacts of Cigarette Tax Reductions on Youth Smoking in Canada

The Economic Impacts of Cigarette Tax Reductions on Youth Smoking in Canada The Economc Impacts of Cgarette Tax Reductons on Youth Smokng n Canada Dane P. Dupont and Anthony J. Ward Economcs, Brock Unversty December 2002 Abstract Cgarettes are the most commonly consumed recreatonal

More information

Location Factors for Non-Ferrous Exploration Investments

Location Factors for Non-Ferrous Exploration Investments Locaton Factors for Non-Ferrous Exploraton Investments Irna Khndanova Unversty of Denver Ths paper analyzes the relatve mportance of geologcal potental and nvestment clmate for nonferrous mnerals exploraton

More information

Trade Adjustment and Productivity in Large Crises. Online Appendix May 2013. Appendix A: Derivation of Equations for Productivity

Trade Adjustment and Productivity in Large Crises. Online Appendix May 2013. Appendix A: Derivation of Equations for Productivity Trade Adjustment Productvty n Large Crses Gta Gopnath Department of Economcs Harvard Unversty NBER Brent Neman Booth School of Busness Unversty of Chcago NBER Onlne Appendx May 2013 Appendx A: Dervaton

More information

Staff Paper. Farm Savings Accounts: Examining Income Variability, Eligibility, and Benefits. Brent Gloy, Eddy LaDue, and Charles Cuykendall

Staff Paper. Farm Savings Accounts: Examining Income Variability, Eligibility, and Benefits. Brent Gloy, Eddy LaDue, and Charles Cuykendall SP 2005-02 August 2005 Staff Paper Department of Appled Economcs and Management Cornell Unversty, Ithaca, New York 14853-7801 USA Farm Savngs Accounts: Examnng Income Varablty, Elgblty, and Benefts Brent

More information

Survive Then Thrive: Determinants of Success in the Economics Ph.D. Program. Wayne A. Grove Le Moyne College, Economics Department

Survive Then Thrive: Determinants of Success in the Economics Ph.D. Program. Wayne A. Grove Le Moyne College, Economics Department Survve Then Thrve: Determnants of Success n the Economcs Ph.D. Program Wayne A. Grove Le Moyne College, Economcs Department Donald H. Dutkowsky Syracuse Unversty, Economcs Department Andrew Grodner East

More information

Is Thailand s Fiscal System Pro-Poor?: Looking from Income and Expenditure Components. Hyun Hwa Son

Is Thailand s Fiscal System Pro-Poor?: Looking from Income and Expenditure Components. Hyun Hwa Son Is Thaland s Fscal System Pro-Poor?: Loong from Income and Expendture Components Hyun Hwa Son The World Ban 88 H Street, NW Washngton, D.C. 20433, U.S.A. Emal: hson@worldban.org Abstract: Ths paper develops

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

Underwriting Risk. Glenn Meyers. Insurance Services Office, Inc.

Underwriting Risk. Glenn Meyers. Insurance Services Office, Inc. Underwrtng Rsk By Glenn Meyers Insurance Servces Offce, Inc. Abstract In a compettve nsurance market, nsurers have lmted nfluence on the premum charged for an nsurance contract. hey must decde whether

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

AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE

AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE Yu-L Huang Industral Engneerng Department New Mexco State Unversty Las Cruces, New Mexco 88003, U.S.A. Abstract Patent

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

Management Quality, Financial and Investment Policies, and. Asymmetric Information

Management Quality, Financial and Investment Policies, and. Asymmetric Information Management Qualty, Fnancal and Investment Polces, and Asymmetrc Informaton Thomas J. Chemmanur * Imants Paegls ** and Karen Smonyan *** Current verson: December 2007 * Professor of Fnance, Carroll School

More information

Dynamics of Toursm Demand Models in Japan

Dynamics of Toursm Demand Models in Japan hort-run and ong-run structural nternatonal toursm demand modelng based on Dynamc AID model -An emprcal research n Japan- Atsush KOIKE a, Dasuke YOHINO b a Graduate chool of Engneerng, Kobe Unversty, Kobe,

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

LIFETIME INCOME OPTIONS

LIFETIME INCOME OPTIONS LIFETIME INCOME OPTIONS May 2011 by: Marca S. Wagner, Esq. The Wagner Law Group A Professonal Corporaton 99 Summer Street, 13 th Floor Boston, MA 02110 Tel: (617) 357-5200 Fax: (617) 357-5250 www.ersa-lawyers.com

More information

Education and Family Income

Education and Family Income Prelmnary: Comments Welcome Educaton and Famly Income Jo Blanden*, Paul Gregg** and Stephen Machn* May 2002 * Department of Economcs, Unversty College London and Centre for Economc Performance, London

More information

Military Conscription and University Enrolment: Evidence from Italy

Military Conscription and University Enrolment: Evidence from Italy DISCUSSION PAPER SERIES IZA DP No. 4212 Mltary Conscrpton and Unversty Enrolment: Evdence from Italy Gorgo D Petro June 2009 Forschungsnsttut zur Zukunft der Arbet Insttute for the Study of Labor Mltary

More information

IDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS

IDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS IDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS Chrs Deeley* Last revsed: September 22, 200 * Chrs Deeley s a Senor Lecturer n the School of Accountng, Charles Sturt Unversty,

More information

When Talk is Free : The Effect of Tariff Structure on Usage under Two- and Three-Part Tariffs

When Talk is Free : The Effect of Tariff Structure on Usage under Two- and Three-Part Tariffs 0 When Talk s Free : The Effect of Tarff Structure on Usage under Two- and Three-Part Tarffs Eva Ascarza Ana Lambrecht Naufel Vlcassm July 2012 (Forthcomng at Journal of Marketng Research) Eva Ascarza

More information

Using an Ordered Probit Regression Model to Assess the Performance of Real Estate Brokers

Using an Ordered Probit Regression Model to Assess the Performance of Real Estate Brokers Usng an Ordered Probt Regresson Model to Assess the Performance of Real Estate Brokers Chun-Chang Lee, Department of Real Estate Management, Natonal Pngtung Insttute of Commerce, Tawan Shu-Man You, Department

More information

Menno Pradhan 1,2,3. Fadia Saadah 1. Robert Sparrow 2. (Revised version September 2004)

Menno Pradhan 1,2,3. Fadia Saadah 1. Robert Sparrow 2. (Revised version September 2004) TI 2003-016/2 Tnbergen Insttute Dscusson Paper Dd the Healthcard Program ensure Access to Medcal Care for the Poor durng Indonesa's Economc Crss? (Revsed verson September 2004) Menno Pradhan 1,2,3 Fada

More information

The Current Employment Statistics (CES) survey,

The Current Employment Statistics (CES) survey, Busness Brths and Deaths Impact of busness brths and deaths n the payroll survey The CES probablty-based sample redesgn accounts for most busness brth employment through the mputaton of busness deaths,

More information

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.

More information

Wage inequality and returns to schooling in Europe: a semi-parametric approach using EU-SILC data

Wage inequality and returns to schooling in Europe: a semi-parametric approach using EU-SILC data MPRA Munch Personal RePEc Archve Wage nequalty and returns to schoolng n Europe: a sem-parametrc approach usng EU-SILC data Marco Bagett and Sergo Sccchtano Unversty La Sapenza Rome, Mnstry of Economc

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

An Empirical Study of Search Engine Advertising Effectiveness

An Empirical Study of Search Engine Advertising Effectiveness An Emprcal Study of Search Engne Advertsng Effectveness Sanjog Msra, Smon School of Busness Unversty of Rochester Edeal Pnker, Smon School of Busness Unversty of Rochester Alan Rmm-Kaufman, Rmm-Kaufman

More information

Management Quality and Equity Issue Characteristics: A Comparison of SEOs and IPOs

Management Quality and Equity Issue Characteristics: A Comparison of SEOs and IPOs Management Qualty and Equty Issue Characterstcs: A Comparson of SEOs and IPOs Thomas J. Chemmanur * Imants Paegls ** and Karen Smonyan *** Current verson: November 2009 (Accepted, Fnancal Management, February

More information

THE EFFECT OF PREPAYMENT PENALTIES ON THE PRICING OF SUBPRIME MORTGAGES

THE EFFECT OF PREPAYMENT PENALTIES ON THE PRICING OF SUBPRIME MORTGAGES THE EFFECT OF PREPAYMENT PENALTIES ON THE PRICING OF SUBPRIME MORTGAGES Gregory Ellehausen, Fnancal Servces Research Program George Washngton Unversty Mchael E. Staten, Fnancal Servces Research Program

More information

The Probability of Informed Trading and the Performance of Stock in an Order-Driven Market

The Probability of Informed Trading and the Performance of Stock in an Order-Driven Market Asa-Pacfc Journal of Fnancal Studes (2007) v36 n6 pp871-896 The Probablty of Informed Tradng and the Performance of Stock n an Order-Drven Market Ta Ma * Natonal Sun Yat-Sen Unversty, Tawan Mng-hua Hseh

More information