Direct and indirect effects in a logit model
|
|
- Joan June Snow
- 8 years ago
- Views:
Transcription
1 Department of Social Research Methodology Vrije Universiteit Amsterdam
2 Outline The aim
3 The Total Effect X Y
4 The Total Effect parental class attend college
5 The Indirect Effect Z a b X Y
6 The Indirect Effect during high school a b parental class attend college
7 The Direct Effect Z a b X c Y
8 The Direct Effect during high school a b parental class c attend college
9 The aim Z The aim is to find the size of the indirect effect relative to the total effect. a b X c Y
10 Outline The aim
11 Estimation When using regress: 1. college = class + 2. college = class
12 Estimation When using regress: 1. college = class + 2. college = class The direct effect is the effect of class in model 1.
13 Estimation When using regress: 1. college = class + 2. college = class The direct effect is the effect of class in model 1. The total effect is the effect of class in model 2.
14 Estimation When using regress: 1. college = class + 2. college = class The direct effect is the effect of class in model 1. The total effect is the effect of class in model 2. The indirect effect is the total effect - direct effect.
15 Estimation When using regress: 1. college = class + 2. college = class The direct effect is the effect of class in model 1. The total effect is the effect of class in model 2. The indirect effect is the total effect - direct effect. This won t work when using logit
16 Why the naive method doesn t work Easiest explained when there is no indirect effect.
17 Why the naive method doesn t work Easiest explained when there is no indirect effect. The total effect = the direct effect + the indirect effect.
18 Why the naive method doesn t work Easiest explained when there is no indirect effect. The total effect = the direct effect + the indirect effect. So, the total effect should be the same as the direct effect when there is no indirect effect.
19 Why the naive method doesn t work Easiest explained when there is no indirect effect. The total effect = the direct effect + the indirect effect. So, the total effect should be the same as the direct effect when there is no indirect effect. So, the effect of class in a model that controls for (the direct effect ) should be the same as the effect of class in a model that does not control for (the total effect ).
20 Effect while controlling for log odds proportion high medium low 1.5 transformation controlled 3 proportion high status log odds log odds low status proportion effect controlled
21 Averaging the proportions 3 log odds proportion high medium low not controlled 1.5 transformation controlled 3 proportion high status log odds log odds low status proportion effect controlled
22 Effect while not controlling for 3 log odds proportion high status log odds log odds low status proportion proportion high medium low not controlled transformation controlled not constrolled effect controlled not constrolled
23 Outline The aim
24 Indirect effect present log odds prop. high status log odds log odds low status prop proportion high medium low not controlled transformation controlled not constrolled effect controlled not constrolled
25 Indirect effect 3 log odds prop. high status log odds indirect effect log odds low status prop proportion factual high medium low not controlled counterfactual high low not controlled
26 Direct effect 3 log odds prop. high status log odds direct effect log odds low status prop proportion factual high medium low not controlled counterfactual high low not controlled
27 Direct and indirect effects in logit 3 log odds proportion high status log odds indirect effect direct effect total effect log odds low status proportion proportion factual high medium low not controlled counterfactual high low not controlled
28 The logic can be reversed 3 log odds total effect direct effect proportion factual high medium low not controlled prop. high status log odds indirect effect log odds low status prop. counterfactual high low not constrolled
29 Extension Erikson et al. (2005) propose to compute the average proportions given the observed and counterfactual distribution of by assuming that is normally distributed, and then integrate over this normal distribution.
30 Extension Erikson et al. (2005) propose to compute the average proportions given the observed and counterfactual distribution of by assuming that is normally distributed, and then integrate over this normal distribution. Alternatively, these averages can be computed by predicting the observed and counterfactual proportions, add them up and divide by the number of respondents in that group.
31 Extension Erikson et al. (2005) propose to compute the average proportions given the observed and counterfactual distribution of by assuming that is normally distributed, and then integrate over this normal distribution. Alternatively, these averages can be computed by predicting the observed and counterfactual proportions, add them up and divide by the number of respondents in that group. The latter method has the advantage of making less assumptions about the distribution of, as it integrates over the empirical distribution of instead of over a normal distribution.
32 Outline The aim
33 Descriptives. table ocf57 if!missing(hsrankq, college), /// > contents(mean college mean hsrankq freq) /// > format(%9.3g) stubwidth(15) occupation of r father in 1957 mean(college) mean(hsrankq) Freq. lower ,218 middle higher ,837
34 The ldecomp package ldecomp depvar [ if ] [ in ] [ weight ], direct(varname) indirect(varlist) [ obspr predpr predodds or rindirect normal range(##) nip(#) interactions nolegend nodecomp nobootstrap bootstrap_options ]
35 Decomposition of log odds ratios. ldecomp college, direct(ocf57) indirect(hsrankq) rind nolegend (running _ldecomp on estimation sample) Bootstrap replications (50) Bootstrap results Number of obs = 8923 Replications = 50 Observed Bootstrap Normal-based Coef. Std. Err. z P> z [95% Conf. Interval] 2/1 total indirect direct indirect direct /1 total indirect direct indirect direct /2 total indirect direct indirect direct
36 Relative effects 2/1r 3/1r 3/2r method method average method method average method method average
37 Decomposition of odds ratios. ldecomp college, direct(ocf57) indirect(hsrankq) or nolegend (running _ldecomp on estimation sample) Bootstrap replications (50) Bootstrap results Number of obs = 8923 Replications = 50 Observed Bootstrap Normal-based Odds Ratio Std. Err. z P> z [95% Conf. Interval] 2/1 total indirect direct indirect direct /1 total indirect direct indirect direct /2 total indirect direct indirect direct
38 Does it matter? Table: Comparing different estimates of the size of indirect effect relative to the size of the total effect generalization (Erikson et al. 2005) naive middle v. low method method average high v. low method method average high v. middle method method average
39 Discussion This is an area of active research
40 Discussion There are unanswered questions:
41 Discussion There are unanswered questions: The need to take the average indirect effect is less than elegant.
42 Discussion There are unanswered questions: The need to take the average indirect effect is less than elegant. How does it relate to the alternative method proposed by Fairlie (2005) and implemented by Ben Jann as the fairlie package?
43 References Buis, M. L.. Erikson, R., J. H. Goldthorpe, M. Jackson, M. Yaish, and D. R. Cox. On class differentials in educational attainment. Proceedings of the National Academy of Science, 102: , Fairlie, R. W. An extension of the Blinder-Oaxaca decomposition technique to logit and probit models. Journal of Economic and Social Measurement, 30: , 2005.
Institut für Soziologie Eberhard Karls Universität Tübingen www.maartenbuis.nl
from Indirect Extracting from Institut für Soziologie Eberhard Karls Universität Tübingen www.maartenbuis.nl from Indirect What is the effect of x on y? Which effect do I choose: average marginal or marginal
More informationESTIMATING AVERAGE TREATMENT EFFECTS: IV AND CONTROL FUNCTIONS, II Jeff Wooldridge Michigan State University BGSE/IZA Course in Microeconometrics
ESTIMATING AVERAGE TREATMENT EFFECTS: IV AND CONTROL FUNCTIONS, II Jeff Wooldridge Michigan State University BGSE/IZA Course in Microeconometrics July 2009 1. Quantile Treatment Effects 2. Control Functions
More informationModule 14: Missing Data Stata Practical
Module 14: Missing Data Stata Practical Jonathan Bartlett & James Carpenter London School of Hygiene & Tropical Medicine www.missingdata.org.uk Supported by ESRC grant RES 189-25-0103 and MRC grant G0900724
More informationCompeting-risks regression
Competing-risks regression Roberto G. Gutierrez Director of Statistics StataCorp LP Stata Conference Boston 2010 R. Gutierrez (StataCorp) Competing-risks regression July 15-16, 2010 1 / 26 Outline 1. Overview
More informationFrom this it is not clear what sort of variable that insure is so list the first 10 observations.
MNL in Stata We have data on the type of health insurance available to 616 psychologically depressed subjects in the United States (Tarlov et al. 1989, JAMA; Wells et al. 1989, JAMA). The insurance is
More informationThe Stata Journal. Editor Nicholas J. Cox Department of Geography Durham University South Road Durham City DH1 3LE UK n.j.cox@stata-journal.
The Stata Journal Editor H. Joseph Newton Department of Statistics Texas A&M University College Station, Texas 77843 979-845-8817; fax 979-845-6077 jnewton@stata-journal.com Associate Editors Christopher
More informationEstimating the random coefficients logit model of demand using aggregate data
Estimating the random coefficients logit model of demand using aggregate data David Vincent Deloitte Economic Consulting London, UK davivincent@deloitte.co.uk September 14, 2012 Introduction Estimation
More informationDiscussion Section 4 ECON 139/239 2010 Summer Term II
Discussion Section 4 ECON 139/239 2010 Summer Term II 1. Let s use the CollegeDistance.csv data again. (a) An education advocacy group argues that, on average, a person s educational attainment would increase
More informationMarginal Effects for Continuous Variables Richard Williams, University of Notre Dame, http://www3.nd.edu/~rwilliam/ Last revised February 21, 2015
Marginal Effects for Continuous Variables Richard Williams, University of Notre Dame, http://www3.nd.edu/~rwilliam/ Last revised February 21, 2015 References: Long 1997, Long and Freese 2003 & 2006 & 2014,
More informationFailure to take the sampling scheme into account can lead to inaccurate point estimates and/or flawed estimates of the standard errors.
Analyzing Complex Survey Data: Some key issues to be aware of Richard Williams, University of Notre Dame, http://www3.nd.edu/~rwilliam/ Last revised January 24, 2015 Rather than repeat material that is
More informationMultinomial and Ordinal Logistic Regression
Multinomial and Ordinal Logistic Regression ME104: Linear Regression Analysis Kenneth Benoit August 22, 2012 Regression with categorical dependent variables When the dependent variable is categorical,
More informationImplementation Committee for Gender Based Salary Adjustments (as identified in the Pay Equity Report, 2005)
Implementation Committee for Gender Based Salary Adjustments (as identified in the Pay Equity Report, 2005) Final Report March 2006 Implementation Committee for Gender Based Salary Adjustments (as identified
More informationHURDLE AND SELECTION MODELS Jeff Wooldridge Michigan State University BGSE/IZA Course in Microeconometrics July 2009
HURDLE AND SELECTION MODELS Jeff Wooldridge Michigan State University BGSE/IZA Course in Microeconometrics July 2009 1. Introduction 2. A General Formulation 3. Truncated Normal Hurdle Model 4. Lognormal
More informationImplementation Committee for Gender Based Salary Adjustments (as identified in the Pay Equity Report, 2005)
Implementation Committee for Gender Based Salary Adjustments (as identified in the Pay Equity Report, 2005) Final Report March 2006 Implementation Committee for Gender Based Salary Adjustments (as identified
More informationFOREIGN AFFAIRS PROGRAM EVALUATION GLOSSARY CORE TERMS
Activity: A specific action or process undertaken over a specific period of time by an organization to convert resources to products or services to achieve results. Related term: Project. Appraisal: An
More informationAn assessment of consumer willingness to pay for Renewable Energy Sources use in Italy: a payment card approach.
An assessment of consumer willingness to pay for Renewable Energy Sources use in Italy: a payment card approach. -First findings- University of Perugia Department of Economics, Finance and Statistics 1
More informationAugust 2012 EXAMINATIONS Solution Part I
August 01 EXAMINATIONS Solution Part I (1) In a random sample of 600 eligible voters, the probability that less than 38% will be in favour of this policy is closest to (B) () In a large random sample,
More informationDepartment of Economics Session 2012/2013. EC352 Econometric Methods. Solutions to Exercises from Week 10 + 0.0077 (0.052)
Department of Economics Session 2012/2013 University of Essex Spring Term Dr Gordon Kemp EC352 Econometric Methods Solutions to Exercises from Week 10 1 Problem 13.7 This exercise refers back to Equation
More informationInteraction effects between continuous variables (Optional)
Interaction effects between continuous variables (Optional) Richard Williams, University of Notre Dame, http://www.nd.edu/~rwilliam/ Last revised February 0, 05 This is a very brief overview of this somewhat
More informationCalculating Effect-Sizes
Calculating Effect-Sizes David B. Wilson, PhD George Mason University August 2011 The Heart and Soul of Meta-analysis: The Effect Size Meta-analysis shifts focus from statistical significance to the direction
More informationRegression III: Advanced Methods
Lecture 4: Transformations Regression III: Advanced Methods William G. Jacoby Michigan State University Goals of the lecture The Ladder of Roots and Powers Changing the shape of distributions Transforming
More informationHealth Insurance Decisions, Expectations, and Job Turnover. Randall P. Ellis Boston University and UTS-CHERE Albert Ma Boston University
Health Insurance Decisions, Expectations, and Job Turnover Randall P. Ellis Boston University and UTS-CHERE Albert Ma Boston University Outline of presentation Introduction Policy context and prior literature
More informationMissing Data & How to Deal: An overview of missing data. Melissa Humphries Population Research Center
Missing Data & How to Deal: An overview of missing data Melissa Humphries Population Research Center Goals Discuss ways to evaluate and understand missing data Discuss common missing data methods Know
More informationOrdinal Regression. Chapter
Ordinal Regression Chapter 4 Many variables of interest are ordinal. That is, you can rank the values, but the real distance between categories is unknown. Diseases are graded on scales from least severe
More informationSAS Software to Fit the Generalized Linear Model
SAS Software to Fit the Generalized Linear Model Gordon Johnston, SAS Institute Inc., Cary, NC Abstract In recent years, the class of generalized linear models has gained popularity as a statistical modeling
More informationCorporate Defaults and Large Macroeconomic Shocks
Corporate Defaults and Large Macroeconomic Shocks Mathias Drehmann Bank of England Andrew Patton London School of Economics and Bank of England Steffen Sorensen Bank of England The presentation expresses
More informationNested Logit. Brad Jones 1. April 30, 2008. University of California, Davis. 1 Department of Political Science. POL 213: Research Methods
Nested Logit Brad 1 1 Department of Political Science University of California, Davis April 30, 2008 Nested Logit Interesting model that does not have IIA property. Possible candidate model for structured
More informationProbit Analysis By: Kim Vincent
Probit Analysis By: Kim Vincent Quick Overview Probit analysis is a type of regression used to analyze binomial response variables. It transforms the sigmoid dose-response curve to a straight line that
More informationEditor Executive Editor Associate Editors Copyright Statement:
The Stata Journal Editor H. Joseph Newton Department of Statistics Texas A & M University College Station, Texas 77843 979-845-3142 979-845-3144 FAX jnewton@stata-journal.com Associate Editors Christopher
More informationNonlinear Regression Functions. SW Ch 8 1/54/
Nonlinear Regression Functions SW Ch 8 1/54/ The TestScore STR relation looks linear (maybe) SW Ch 8 2/54/ But the TestScore Income relation looks nonlinear... SW Ch 8 3/54/ Nonlinear Regression General
More informationIII. INTRODUCTION TO LOGISTIC REGRESSION. a) Example: APACHE II Score and Mortality in Sepsis
III. INTRODUCTION TO LOGISTIC REGRESSION 1. Simple Logistic Regression a) Example: APACHE II Score and Mortality in Sepsis The following figure shows 30 day mortality in a sample of septic patients as
More informationLOGIT AND PROBIT ANALYSIS
LOGIT AND PROBIT ANALYSIS A.K. Vasisht I.A.S.R.I., Library Avenue, New Delhi 110 012 amitvasisht@iasri.res.in In dummy regression variable models, it is assumed implicitly that the dependent variable Y
More informationMULTIPLE REGRESSION EXAMPLE
MULTIPLE REGRESSION EXAMPLE For a sample of n = 166 college students, the following variables were measured: Y = height X 1 = mother s height ( momheight ) X 2 = father s height ( dadheight ) X 3 = 1 if
More informationSUMAN DUVVURU STAT 567 PROJECT REPORT
SUMAN DUVVURU STAT 567 PROJECT REPORT SURVIVAL ANALYSIS OF HEROIN ADDICTS Background and introduction: Current illicit drug use among teens is continuing to increase in many countries around the world.
More informationSample Size Calculation for Longitudinal Studies
Sample Size Calculation for Longitudinal Studies Phil Schumm Department of Health Studies University of Chicago August 23, 2004 (Supported by National Institute on Aging grant P01 AG18911-01A1) Introduction
More informationFrom the help desk: hurdle models
The Stata Journal (2003) 3, Number 2, pp. 178 184 From the help desk: hurdle models Allen McDowell Stata Corporation Abstract. This article demonstrates that, although there is no command in Stata for
More informationHow to set the main menu of STATA to default factory settings standards
University of Pretoria Data analysis for evaluation studies Examples in STATA version 11 List of data sets b1.dta (To be created by students in class) fp1.xls (To be provided to students) fp1.txt (To be
More informationMODEL I: DRINK REGRESSED ON GPA & MALE, WITHOUT CENTERING
Interpreting Interaction Effects; Interaction Effects and Centering Richard Williams, University of Notre Dame, http://www3.nd.edu/~rwilliam/ Last revised February 20, 2015 Models with interaction effects
More informationThe Stata Journal. Editor Nicholas J. Cox Department of Geography Durham University South Road Durham City DH1 3LE UK n.j.cox@stata-journal.
The Stata Journal Editor H. Joseph Newton Department of Statistics Texas A&M University College Station, Texas 77843 979-845-8817; fax 979-845-6077 jnewton@stata-journal.com Associate Editors Christopher
More informationIntroduction. Survival Analysis. Censoring. Plan of Talk
Survival Analysis Mark Lunt Arthritis Research UK Centre for Excellence in Epidemiology University of Manchester 01/12/2015 Survival Analysis is concerned with the length of time before an event occurs.
More informationStatistical modelling with missing data using multiple imputation. Session 4: Sensitivity Analysis after Multiple Imputation
Statistical modelling with missing data using multiple imputation Session 4: Sensitivity Analysis after Multiple Imputation James Carpenter London School of Hygiene & Tropical Medicine Email: james.carpenter@lshtm.ac.uk
More informationis paramount in advancing any economy. For developed countries such as
Introduction The provision of appropriate incentives to attract workers to the health industry is paramount in advancing any economy. For developed countries such as Australia, the increasing demand for
More information10 Dichotomous or binary responses
10 Dichotomous or binary responses 10.1 Introduction Dichotomous or binary responses are widespread. Examples include being dead or alive, agreeing or disagreeing with a statement, and succeeding or failing
More informationCorrelation and Regression
Correlation and Regression Scatterplots Correlation Explanatory and response variables Simple linear regression General Principles of Data Analysis First plot the data, then add numerical summaries Look
More informationSimple Random Sampling
Source: Frerichs, R.R. Rapid Surveys (unpublished), 2008. NOT FOR COMMERCIAL DISTRIBUTION 3 Simple Random Sampling 3.1 INTRODUCTION Everyone mentions simple random sampling, but few use this method for
More informationWhat Every Employment Lawyer Should Know About Statistical Proof
I Was Told There Would Be No Math: What Every Employment Lawyer Should Know About Statistical Proof In Employment Matters Susan E. Dunnings Vice President, Associate General Counsel Lockheed Martin Corporation
More informationLinear Regression Models with Logarithmic Transformations
Linear Regression Models with Logarithmic Transformations Kenneth Benoit Methodology Institute London School of Economics kbenoit@lse.ac.uk March 17, 2011 1 Logarithmic transformations of variables Considering
More informationUnit 12 Logistic Regression Supplementary Chapter 14 in IPS On CD (Chap 16, 5th ed.)
Unit 12 Logistic Regression Supplementary Chapter 14 in IPS On CD (Chap 16, 5th ed.) Logistic regression generalizes methods for 2-way tables Adds capability studying several predictors, but Limited to
More informationThe CRM for ordinal and multivariate outcomes. Elizabeth Garrett-Mayer, PhD Emily Van Meter
The CRM for ordinal and multivariate outcomes Elizabeth Garrett-Mayer, PhD Emily Van Meter Hollings Cancer Center Medical University of South Carolina Outline Part 1: Ordinal toxicity model Part 2: Efficacy
More informationLogistic Regression. BUS 735: Business Decision Making and Research
Goals of this section 2/ 8 Specific goals: Learn how to conduct regression analysis with a dummy independent variable. Learning objectives: LO2: Be able to construct and use multiple regression models
More informationSimple Linear Regression
STAT 101 Dr. Kari Lock Morgan Simple Linear Regression SECTIONS 9.3 Confidence and prediction intervals (9.3) Conditions for inference (9.1) Want More Stats??? If you have enjoyed learning how to analyze
More informationPrivate Forms of Unemployment Protection and Social Stratification. Alison Koslowski University of Edinburgh, UK
Private Forms of Unemployment Protection and Social Stratification in England and Scotland Alison Koslowski University of Edinburgh, UK Welfare Markets and Personal Risk Management in England and Scotland
More informationGLOSSARY OF EVALUATION TERMS
Planning and Performance Management Unit Office of the Director of U.S. Foreign Assistance Final Version: March 25, 2009 INTRODUCTION This Glossary of Evaluation and Related Terms was jointly prepared
More informationSoftware for Analysis of YRBS Data
Youth Risk Behavior Surveillance System (YRBSS) Software for Analysis of YRBS Data June 2014 Where can I get more information? Visit www.cdc.gov/yrbss or call 800 CDC INFO (800 232 4636). CONTENTS Overview
More informationand Gologit2: A Program for Ordinal Variables Last revised May 12, 2005 Page 1 ologit y x1 x2 x3 gologit2 y x1 x2 x3, pl lrforce
Gologit2: A Program for Generalized Logistic Regression/ Partial Proportional Odds Models for Ordinal Dependent Variables Richard Williams, Richard.A.Williams.5@ND.Edu Last revised May 12, 2005 [This document
More informationBasic Statistical and Modeling Procedures Using SAS
Basic Statistical and Modeling Procedures Using SAS One-Sample Tests The statistical procedures illustrated in this handout use two datasets. The first, Pulse, has information collected in a classroom
More informationWealth inequality: Britain in international perspective. Frank Cowell: Wealth Seminar June 2012
Wealth inequality: Britain in international perspective Frank Cowell: Wealth Seminar June 2012 Questions What does UK wealth inequality look like in context? What is role of inequality among the rich?
More informationThe Regression Calibration Method for Fitting Generalized Linear Models with Additive Measurement Error
The Stata Journal (), Number, pp. 1 11 The Regression Calibration Method for Fitting Generalized Linear Models with Additive Measurement Error James W. Hardin Norman J. Arnold School of Public Health University
More informationECON 142 SKETCH OF SOLUTIONS FOR APPLIED EXERCISE #2
University of California, Berkeley Prof. Ken Chay Department of Economics Fall Semester, 005 ECON 14 SKETCH OF SOLUTIONS FOR APPLIED EXERCISE # Question 1: a. Below are the scatter plots of hourly wages
More informationIn the general population of 0 to 4-year-olds, the annual incidence of asthma is 1.4%
Hypothesis Testing for a Proportion Example: We are interested in the probability of developing asthma over a given one-year period for children 0 to 4 years of age whose mothers smoke in the home In the
More informationStatistics 104 Final Project A Culture of Debt: A Study of Credit Card Spending in America TF: Kevin Rader Anonymous Students: LD, MH, IW, MY
Statistics 104 Final Project A Culture of Debt: A Study of Credit Card Spending in America TF: Kevin Rader Anonymous Students: LD, MH, IW, MY ABSTRACT: This project attempted to determine the relationship
More informationHealth inequality and the economic crisis: what do we know?
Health inequality and the economic crisis: what do we know? Eddy van Doorslaer Professor of Health Economics Erasmus School of Economics Erasmus University Rotterdam (with Pilar Garcia Gomez and Tom van
More informationDevelopment of the nomolog program and its evolution
Development of the nomolog program and its evolution Towards the implementation of a nomogram generator for the Cox regression Alexander Zlotnik, Telecom.Eng. Víctor Abraira Santos, PhD Ramón y Cajal University
More informationRegression Modeling Strategies
Frank E. Harrell, Jr. Regression Modeling Strategies With Applications to Linear Models, Logistic Regression, and Survival Analysis With 141 Figures Springer Contents Preface Typographical Conventions
More informationIt is important to bear in mind that one of the first three subscripts is redundant since k = i -j +3.
IDENTIFICATION AND ESTIMATION OF AGE, PERIOD AND COHORT EFFECTS IN THE ANALYSIS OF DISCRETE ARCHIVAL DATA Stephen E. Fienberg, University of Minnesota William M. Mason, University of Michigan 1. INTRODUCTION
More informationIAPRI Quantitative Analysis Capacity Building Series. Multiple regression analysis & interpreting results
IAPRI Quantitative Analysis Capacity Building Series Multiple regression analysis & interpreting results How important is R-squared? R-squared Published in Agricultural Economics 0.45 Best article of the
More informationChapter 18. Effect modification and interactions. 18.1 Modeling effect modification
Chapter 18 Effect modification and interactions 18.1 Modeling effect modification weight 40 50 60 70 80 90 100 male female 40 50 60 70 80 90 100 male female 30 40 50 70 dose 30 40 50 70 dose Figure 18.1:
More informationespecially with continuous
Handling interactions in Stata, especially with continuous predictors Patrick Royston & Willi Sauerbrei German Stata Users meeting, Berlin, 1 June 2012 Interactions general concepts General idea of a (two-way)
More informationMarginal Person. Average Person. (Average Return of College Goers) Return, Cost. (Average Return in the Population) (Marginal Return)
1 2 3 Marginal Person Average Person (Average Return of College Goers) Return, Cost (Average Return in the Population) 4 (Marginal Return) 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
More informationOctober 31, 2014. The effect of price on youth alcohol. consumption in Canada. Stephenson Strobel & Evelyn Forget. Introduction. Data and Methodology
October 31, 2014 Why should we care? Deaths 0 500 1000 1500 2000 2500 2000 2002 2004 2006 2008 2010 2012 Year Accidents (unintentional injuries) Intentional self-harm (suicide) Other causes of death Assault
More informationLecture 1: Review and Exploratory Data Analysis (EDA)
Lecture 1: Review and Exploratory Data Analysis (EDA) Sandy Eckel seckel@jhsph.edu Department of Biostatistics, The Johns Hopkins University, Baltimore USA 21 April 2008 1 / 40 Course Information I Course
More informationLogit and Probit. Brad Jones 1. April 21, 2009. University of California, Davis. Bradford S. Jones, UC-Davis, Dept. of Political Science
Logit and Probit Brad 1 1 Department of Political Science University of California, Davis April 21, 2009 Logit, redux Logit resolves the functional form problem (in terms of the response function in the
More informationAddressing Alternative. Multiple Regression. 17.871 Spring 2012
Addressing Alternative Explanations: Multiple Regression 17.871 Spring 2012 1 Did Clinton hurt Gore example Did Clinton hurt Gore in the 2000 election? Treatment is not liking Bill Clinton 2 Bivariate
More informationIntroduction to structural equation modeling using the sem command
Introduction to structural equation modeling using the sem command Gustavo Sanchez Senior Econometrician StataCorp LP Mexico City, Mexico Gustavo Sanchez (StataCorp) November 13, 2014 1 / 33 Outline Outline
More informationEducation and Wage Differential by Race: Convergence or Divergence? *
Education and Wage Differential by Race: Convergence or Divergence? * Tian Luo Thesis Advisor: Professor Andrea Weber University of California, Berkeley Department of Economics April 2009 Abstract This
More informationLinda K. Muthén Bengt Muthén. Copyright 2008 Muthén & Muthén www.statmodel.com. Table Of Contents
Mplus Short Courses Topic 2 Regression Analysis, Eploratory Factor Analysis, Confirmatory Factor Analysis, And Structural Equation Modeling For Categorical, Censored, And Count Outcomes Linda K. Muthén
More informationLogistic regression modeling the probability of success
Logistic regression modeling the probability of success Regression models are usually thought of as only being appropriate for target variables that are continuous Is there any situation where we might
More informationClassification Problems
Classification Read Chapter 4 in the text by Bishop, except omit Sections 4.1.6, 4.1.7, 4.2.4, 4.3.3, 4.3.5, 4.3.6, 4.4, and 4.5. Also, review sections 1.5.1, 1.5.2, 1.5.3, and 1.5.4. Classification Problems
More informationUsing Stata for Categorical Data Analysis
Using Stata for Categorical Data Analysis NOTE: These problems make extensive use of Nick Cox s tab_chi, which is actually a collection of routines, and Adrian Mander s ipf command. From within Stata,
More informationReverse Transfer Students and Postsecondary Outcomes: A Potential Opportunity
Reverse Transfer Students and Postsecondary Outcomes: A Potential Opportunity American Educational Research Association Eric Lichtenberger Associate Director for Research Illinois Education Research Council
More informationStandard errors of marginal effects in the heteroskedastic probit model
Standard errors of marginal effects in the heteroskedastic probit model Thomas Cornelißen Discussion Paper No. 320 August 2005 ISSN: 0949 9962 Abstract In non-linear regression models, such as the heteroskedastic
More informationEffects of Recent Carrot-and-Stick Policy Initiatives on Private Health Insurance Coverage in Australia
THE ECONOMIC RECORD, VOL. 81, NO. 254, SEPTEMBER, 2005, 262 272 Effects of Recent Carrot-and-Stick Policy Initiatives on Private Health Insurance Coverage in Australia ALFONS PALANGKARAYA and JONGSAY YONG
More informationStatistics, Data Analysis & Econometrics
Using the LOGISTIC Procedure to Model Responses to Financial Services Direct Marketing David Marsh, Senior Credit Risk Modeler, Canadian Tire Financial Services, Welland, Ontario ABSTRACT It is more important
More informationLecture 15. Endogeneity & Instrumental Variable Estimation
Lecture 15. Endogeneity & Instrumental Variable Estimation Saw that measurement error (on right hand side) means that OLS will be biased (biased toward zero) Potential solution to endogeneity instrumental
More informationSUPPORTING LOGISTICS DECISIONS BY USING COST AND PERFORMANCE MANAGEMENT TOOLS. Zoltán BOKOR. Abstract. 1. Introduction
SUPPORTING LOGISTICS DECISIONS BY USING COST AND PERFORMANCE MANAGEMENT TOOLS Zoltán BOKOR Department of Transport Economics Faculty of Transportation Engineering Budapest University of Technology and
More informationUsing An Ordered Logistic Regression Model with SAS Vartanian: SW 541
Using An Ordered Logistic Regression Model with SAS Vartanian: SW 541 libname in1 >c:\=; Data first; Set in1.extract; A=1; PROC LOGIST OUTEST=DD MAXITER=100 ORDER=DATA; OUTPUT OUT=CC XBETA=XB P=PROB; MODEL
More informationThe average hotel manager recognizes the criticality of forecasting. However, most
Introduction The average hotel manager recognizes the criticality of forecasting. However, most managers are either frustrated by complex models researchers constructed or appalled by the amount of time
More informationI L L I N O I S UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN
Beckman HLM Reading Group: Questions, Answers and Examples Carolyn J. Anderson Department of Educational Psychology I L L I N O I S UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN Linear Algebra Slide 1 of
More informationA General Approach to Variance Estimation under Imputation for Missing Survey Data
A General Approach to Variance Estimation under Imputation for Missing Survey Data J.N.K. Rao Carleton University Ottawa, Canada 1 2 1 Joint work with J.K. Kim at Iowa State University. 2 Workshop on Survey
More informationWOMEN-OWNED BUSINESSES AND ACCESS TO BANK CREDIT: A TIME SERIES PERSPECTIVE
WOMEN-OWNED BUSINESSES AND ACCESS TO BANK CREDIT: A TIME SERIES PERSPECTIVE Monica Zimmerman Treichel Temple University Fox School of Business and Management Strategic Management Department 1810 North
More informationVI. Introduction to Logistic Regression
VI. Introduction to Logistic Regression We turn our attention now to the topic of modeling a categorical outcome as a function of (possibly) several factors. The framework of generalized linear models
More informationPre-Employment Vocational Education and Training in Korea. November 2009
SP DISCUSSION PAPER NO. 0921 Pre-Employment Vocational Education and Training in Korea ChangKyun Chae and Jaeho Chung November 2009 Pre-Employment Vocational Education and Training in Korea ChangKyun Chae
More informationComplex Survey Design Using Stata
Complex Survey Design Using Stata 2010 This document provides a basic overview of how to handle complex survey design using Stata. Probability weighting and compensating for clustered and stratified samples
More informationThe Wage Return to Education: What Hides Behind the Least Squares Bias?
DISCUSSION PAPER SERIES IZA DP No. 8855 The Wage Return to Education: What Hides Behind the Least Squares Bias? Corrado Andini February 2015 Forschungsinstitut zur Zukunft der Arbeit Institute for the
More informationDoes Financial Sophistication Matter in Retirement Preparedness of U.S Households? Evidence from the 2010 Survey of Consumer Finances
Does Financial Sophistication Matter in Retirement Preparedness of U.S Households? Evidence from the 2010 Survey of Consumer Finances Kyoung Tae Kim, The Ohio State University 1 Sherman D. Hanna, The Ohio
More informationRESEARCH. Ethnicity and Degree Attainment. Stijn Broecke and Tom Nicholls Department for Education and Skills. Research Report RW92
RESEARCH Stijn Broecke and Tom Nicholls Department for Education and Skills Research Report RW92 Research Report No RW92 Stijn Broecke and Tom Nicholls Department for Education and Skills Contents Section
More informationMEASURING THE INVENTORY TURNOVER IN DISTRIBUTIVE TRADE
MEASURING THE INVENTORY TURNOVER IN DISTRIBUTIVE TRADE Marijan Karić, Ph.D. Josip Juraj Strossmayer University of Osijek Faculty of Economics in Osijek Gajev trg 7, 31000 Osijek, Croatia Phone: +385 31
More informationOnline Appendix to Are Risk Preferences Stable Across Contexts? Evidence from Insurance Data
Online Appendix to Are Risk Preferences Stable Across Contexts? Evidence from Insurance Data By LEVON BARSEGHYAN, JEFFREY PRINCE, AND JOSHUA C. TEITELBAUM I. Empty Test Intervals Here we discuss the conditions
More informationYouth Risk Behavior Survey (YRBS) Software for Analysis of YRBS Data
Youth Risk Behavior Survey (YRBS) Software for Analysis of YRBS Data CONTENTS Overview 1 Background 1 1. SUDAAN 2 1.1. Analysis capabilities 2 1.2. Data requirements 2 1.3. Variance estimation 2 1.4. Survey
More informationPublic vs. Private: Characteristics of the Companies Backed by Listed Private Equity
Public vs. Private: Characteristics of the Companies Backed by Listed Private Equity M. Sinan Goktan California State University, East Bay Erdem Ucar University of South Florida Listed Private Equity (LPE)
More information