1

Similar documents
Fixed and Random Effects Models: Vartanian, SW 683

Problem Set 10: Panel Data

Please discuss each of the 3 problems on a separate sheet of paper, not just on a separate page!

point estimates, standard errors, testing, and inference for nonlinear combinations

Lecture 3 Linear random intercept models

Microeconometrics (PhD) Problem set 2: Dynamic Panel Data Solutions

Exercices for Applied Econometrics A

Econometrics Homework 4 Solutions

****Lab 4, Feb 4: EDA and OLS and WLS

Description Quick start Menu Syntax Options Remarks and examples Stored results Methods and formulas Acknowledgment References Also see

4 Instrumental Variables Single endogenous variable One continuous instrument. 2

Simultaneous Equations with Error Components. Mike Bronner Marko Ledic Anja Breitwieser

4 Instrumental Variables Single endogenous variable One continuous instrument. 2

Applied Econometrics. Lecture 3: Introduction to Linear Panel Data Models

Exam ECON3150/4150: Introductory Econometrics. 18 May 2016; 09:00h-12.00h.

Problem Set 5 ANSWERS

Econometrics. 8) Instrumental variables

Эконометрика, , 4 модуль Семинар Для Группы Э_Б2015_Э_3 Семинарист О.А.Демидова

Fortin Econ Econometric Review 1. 1 Panel Data Methods Fixed Effects Dummy Variables Regression... 7

Panel Data: Very Brief Overview Richard Williams, University of Notre Dame, Last revised April 6, 2015

Quantitative Methods Final Exam (2017/1)

Monday 7 th Febraury 2005

University of California at Berkeley Fall Introductory Applied Econometrics Final examination. Scores add up to 125 points

CRE METHODS FOR UNBALANCED PANELS Correlated Random Effects Panel Data Models IZA Summer School in Labor Economics May 13-19, 2013 Jeffrey M.

1 The basics of panel data

multilevel modeling: concepts, applications and interpretations

xtdpdqml: Quasi-maximum likelihood estimation of linear dynamic short-t panel data models

ECON 497 Final Exam Page 1 of 12

Outline. Linear OLS Models vs: Linear Marginal Models Linear Conditional Models. Random Intercepts Random Intercepts & Slopes

Jeffrey M. Wooldridge Michigan State University

Dynamic Panel Data Models

Longitudinal Data Analysis. RatSWD Nachwuchsworkshop Vorlesung von Josef Brüderl 25. August, 2009

Empirical Application of Panel Data Regression

Confidence intervals for the variance component of random-effects linear models

Practice 2SLS with Artificial Data Part 1

Maria Elena Bontempi Roberto Golinelli this version: 5 September 2007

INTRODUCTION TO BASIC LINEAR REGRESSION MODEL

STATA Commands for Unobserved Effects Panel Data

Instrumental Variables, Simultaneous and Systems of Equations

Lecture 3.1 Basic Logistic LDA

Problem Set #3-Key. wage Coef. Std. Err. t P> t [95% Conf. Interval]

Lecture#17. Time series III

Advanced Topics in Maximum Likelihood Models for Panel and Time-Series Cross-Section Data 2009 ICPSR Summer Program

Advanced Topics in Maximum Likelihood Models for Panel and Time-Series Cross-Section Data 2006 ICPSR Summer Program

ERSA Training Workshop Lecture 5: Estimation of Binary Choice Models with Panel Data

Lab 07 Introduction to Econometrics

8. Nonstandard standard error issues 8.1. The bias of robust standard errors

Understanding the multinomial-poisson transformation

Lecture#12. Instrumental variables regression Causal parameters III

Problem Set #5-Key Sonoma State University Dr. Cuellar Economics 317- Introduction to Econometrics

Correlation and Simple Linear Regression

Lecture 4: Generalized Linear Mixed Models

2.1. Consider the following production function, known in the literature as the transcendental production function (TPF).

Measurement Error. Often a data set will contain imperfect measures of the data we would ideally like.

Lab 3: Two levels Poisson models (taken from Multilevel and Longitudinal Modeling Using Stata, p )

Lab 11 - Heteroskedasticity

ECON2228 Notes 10. Christopher F Baum. Boston College Economics. cfb (BC Econ) ECON2228 Notes / 48

Panel Threshold Model

Econometrics Midterm Examination Answers

Interpreting coefficients for transformed variables

ECON2228 Notes 10. Christopher F Baum. Boston College Economics. cfb (BC Econ) ECON2228 Notes / 54

Using Stata for a memory saving fixed effects estimation of the three-way error component model

Editor Nicholas J. Cox Department of Geography Durham University South Road Durham City DH1 3LE UK

Econ 836 Final Exam. 2 w N 2 u N 2. 2 v N

Ecmt 675: Econometrics I

Lecture 8: Instrumental Variables Estimation

Handout 11: Measurement Error

New Developments in Econometrics Lecture 11: Difference-in-Differences Estimation

Specification Error: Omitted and Extraneous Variables

Multilevel Modeling (MLM) part 1. Robert Yu

. *DEFINITIONS OF ARTIFICIAL DATA SET. mat m=(12,20,0) /*matrix of means of RHS vars: edu, exp, error*/

Homework Solutions Applied Logistic Regression

Final Exam. 1. Definitions: Briefly Define each of the following terms as they relate to the material covered in class.

Problem set - Selection and Diff-in-Diff

An explanation of Two Stage Least Squares

Question 1 carries a weight of 25%; question 2 carries 25%; question 3 carries 20%; and question 4 carries 30%.

4/9/2014. Outline for Stochastic Frontier Analysis. Stochastic Frontier Production Function. Stochastic Frontier Production Function

Final Exam. Question 1 (20 points) 2 (25 points) 3 (30 points) 4 (25 points) 5 (10 points) 6 (40 points) Total (150 points) Bonus question (10)

Dynamic Panels. Chapter Introduction Autoregressive Model

Recent Developments in Multilevel Modeling

STATISTICS 110/201 PRACTICE FINAL EXAM

Lecture 5. In the last lecture, we covered. This lecture introduces you to

Section Least Squares Regression

Warwick Economics Summer School Topics in Microeconometrics Instrumental Variables Estimation

Soc 63993, Homework #7 Answer Key: Nonlinear effects/ Intro to path analysis

GMM Estimation in Stata

ECON2228 Notes 7. Christopher F Baum. Boston College Economics. cfb (BC Econ) ECON2228 Notes / 41

Analyzing spatial autoregressive models using Stata

Suggested Answers Problem set 4 ECON 60303

Econometrics II Censoring & Truncation. May 5, 2011

Handout 12. Endogeneity & Simultaneous Equation Models

Spatial Regression Models: Identification strategy using STATA TATIANE MENEZES PIMES/UFPE

sociology sociology Scatterplots Quantitative Research Methods: Introduction to correlation and regression Age vs Income

Graduate Econometrics Lecture 4: Heteroskedasticity

Meta-analysis of epidemiological dose-response studies

Problem Set 1 ANSWERS

Heteroskedasticity Example

ECONOMICS AND ECONOMIC METHODS PRELIM EXAM Statistics and Econometrics August 2013

Model and Working Correlation Structure Selection in GEE Analyses of Longitudinal Data

From the help desk: Comparing areas under receiver operating characteristic curves from two or more probit or logit models

Transcription:

i

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

Foreign Liabilities/ Dependen Variabel Rasio Foreign Liabilities Total Debt (1) (2) Konstanta 0.216*** 0.219*** Ekspor/Total Sales -0.064-0.055 Rasio Foreign Asset 0.538*** 1.785*** R-squared 0.044 0.341 N 890 890 Estimator FE FE Haussman Test 0.002 0.002 25

26

Variabel Dependen Rasio Ekuitas Konstanta -2.269 FA*Nilai Tukar -88.553 *** NFA (-1) 12.641 *** Nilai Tukar (-1) -14.133 *** Rasio Derivatif (-1) -0.021 DL (-1) 0.052 *** R-Squared N Estimator Hausman Test 0.455 756.000 FE 0.000 Dependen Variabel Rasio Net Income Rasio EBIT Rasio Operating Rasio Interest Expenditures Coverage (1) (2) (3) (4) Konstanta 0.556 *** 0.65*** -0.679*** -32.568 Rasio Transaksi Derivatif 0 0 0-0.037 Rasio EBIT 1.067 Rasio Ekspor -0.008 0.044 0.047 Rasio Foreign Asset -0.537 *** -0.763*** 0.563*** 1.160 Rasio Foreign Liabilities 0.995 *** 1.125*** -1.01*** -0.763* Rasio Liabilities -1.312 *** -1.702*** 2.214*** Rasio Sales 0.039 0.109*** -0.131*** Changes GM 0.511** R-squared 0.3041 0.2043 0.144 0.1095 N 761 761 761 882 Estimator FE FE FE RE Haussman Test 0.000 0.000 0.000 0.243 27

28

Variabel Level 1 st Difference CVAR -7.112* DERIVATIF_TOT -5.777* FORWARD_TOT -6.379* SWAP_TOT -3.779** OPTION_TOT -21.166* DER_KORP -41.538* FORWARD_KORP -6.0446* SWAP_KORP -2.438-20.56313* OPTION_KORP -41.984* 29

30

31

32

33

34

35

. xtreg rfrli rexs rfras, fe Fixed-effects (within) regression Number of obs = 890 Group variable: id Number of groups = 128 R-sq: within = 0.0211 Obs per group: min = 5 between = 0.0638 avg = 7.0 overall = 0.0439 max = 7 F(2,760) = 8.19 corr(u_i, Xb) = 0.0444 Prob > F = 0.0003 rfrli Coef. Std. Err. t P> t [95% Conf. Interval] rexs -.0635617.07422-0.86 0.392 -.2092622.0821388 rfras.5378449.1353012 3.98 0.000.2722364.8034534 _cons.2161235.0180443 11.98 0.000.1807009.2515462 sigma_u.30185644 sigma_e.30209503 rho.49960495 (fraction of variance due to u_i) F test that all u_i=0: F(127, 760) = 6.37 Prob > F = 0.0000. estimates store fixed. xtreg rfrli rexs rfras, re Random-effects GLS regression Number of obs = 890 Group variable: id Number of groups = 128 R-sq: within = 0.0179 Obs per group: min = 5 between = 0.1298 avg = 7.0 overall = 0.0745 max = 7 Random effects u_i ~ Gaussian Wald chi2(2) = 29.99 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 rfrli Coef. Std. Err. z P> z [95% Conf. Interval] rexs.0650855.0636606 1.02 0.307 -.0596871.189858 rfras.6167226.1158115 5.33 0.000.3897363.8437089 _cons.1955338.0283887 6.89 0.000.139893.2511747 sigma_u.26174858 sigma_e.30209503 rho.42880842 (fraction of variance due to u_i). estimates store random. hausman fixed random Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rexs -.0635617.0650855 -.1286472.0381566 rfras.5378449.6167226 -.0788777.069958 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(2) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 12.37 Prob>chi2 = 0.0021 36

. xtreg rfrtl rexs rfras, fe Fixed-effects (within) regression Number of obs = 890 Group variable: id Number of groups = 128 R-sq: within = 0.3617 Obs per group: min = 5 between = 0.3328 avg = 7.0 overall = 0.3405 max = 7 F(2,760) = 215.37 corr(u_i, Xb) = -0.1406 Prob > F = 0.0000 rfrtl Coef. Std. Err. t P> t [95% Conf. Interval] rexs -.0545604.0472-1.16 0.248 -.1472182.0380974 rfras 1.784908.0860444 20.74 0.000 1.615995 1.953821 _cons.2190851.0114752 19.09 0.000.1965582.2416121 sigma_u.23865096 sigma_e.19211642 rho.60678096 (fraction of variance due to u_i) F test that all u_i=0: F(127, 760) = 9.96 Prob > F = 0.0000. estimates store fixed. xtreg rfrtl rexs rfras, re Random-effects GLS regression Number of obs = 890 Group variable: id Number of groups = 128 R-sq: within = 0.3603 Obs per group: min = 5 between = 0.3481 avg = 7.0 overall = 0.3500 max = 7 Random effects u_i ~ Gaussian Wald chi2(2) = 491.74 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 rfrtl Coef. Std. Err. z P> z [95% Conf. Interval] rexs.0060429.0426402 0.14 0.887 -.0775304.0896162 rfras 1.719443.0776247 22.15 0.000 1.567302 1.871585 _cons.2196433.0219541 10.00 0.000.176614.2626726 sigma_u.21562353 sigma_e.19211642 rho.55746122 (fraction of variance due to u_i). estimates store random. hausman fixed random Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rexs -.0545604.0060429 -.0606033.0202398 rfras 1.784908 1.719443.0654642.037122 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(2) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 12.59 Prob>chi2 = 0.0018 37

. xtreg rderb rexs rdfras rdfrli, fe Fixed-effects (within) regression Number of obs = 758 Group variable: id Number of groups = 128 R-sq: within = 0.8275 Obs per group: min = 4 between = 0.8292 avg = 5.9 overall = 0.8278 max = 6 F(3,627) = 1002.34 corr(u_i, Xb) = 0.0067 Prob > F = 0.0000 rderb Coef. Std. Err. t P> t [95% Conf. Interval] rexs 30.74096 103.5062 0.30 0.767-172.5198 234.0017 rdfras -63.51326 2.172741-29.23 0.000-67.77999-59.24653 rdfrli -13.67903.7283665-18.78 0.000-15.10936-12.2487 _cons 23.53646 18.54739 1.27 0.205-12.88607 59.95898 sigma_u 157.49605 sigma_e 392.88544 rho.13844863 (fraction of variance due to u_i) F test that all u_i=0: F(127, 627) = 0.96 Prob > F = 0.6051. estimates store fixed. xtreg rderb rexs rdfras rdfrli, re Random-effects GLS regression Number of obs = 758 Group variable: id Number of groups = 128 R-sq: within = 0.8274 Obs per group: min = 4 between = 0.8299 avg = 5.9 overall = 0.8279 max = 6 Random effects u_i ~ Gaussian Wald chi2(3) = 3626.56 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 rderb Coef. Std. Err. z P> z [95% Conf. Interval] rexs -1.70622 54.86008-0.03 0.975-109.23 105.8176 rdfras -63.69067 1.974589-32.26 0.000-67.56079-59.82055 rdfrli -13.66867.6617602-20.66 0.000-14.9657-12.37165 _cons 27.30742 15.58008 1.75 0.080-3.228978 57.84383 sigma_u 0 sigma_e 392.88544 rho 0 (fraction of variance due to u_i). estimates store random. hausman fixed random Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rexs 30.74096-1.70622 32.44718 87.77188 rdfras -63.51326-63.69067.1774059.9065336 rdfrli -13.67903-13.66867 -.0103567.3042879 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(3) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 0.20 Prob>chi2 = 0.9783 38

. xtreg rderj rexs rdfras rdfrli, fe Fixed-effects (within) regression Number of obs = 758 Group variable: id Number of groups = 128 R-sq: within = 0.8277 Obs per group: min = 4 between = 0.8298 avg = 5.9 overall = 0.8281 max = 6 F(3,627) = 1003.90 corr(u_i, Xb) = 0.0071 Prob > F = 0.0000 rderj Coef. Std. Err. t P> t [95% Conf. Interval] rexs 29.5331 101.1357 0.29 0.770-169.0727 228.1389 rdfras -62.12105 2.122982-29.26 0.000-66.29007-57.95203 rdfrli -13.37106.7116858-18.79 0.000-14.76864-11.97348 _cons 22.68138 18.12263 1.25 0.211-12.90702 58.26978 sigma_u 153.77417 sigma_e 383.88777 rho.13827041 (fraction of variance due to u_i) F test that all u_i=0: F(127, 627) = 0.96 Prob > F = 0.6087. estimates store fixed. xtreg rderj rexs rdfras rdfrli, re Random-effects GLS regression Number of obs = 758 Group variable: id Number of groups = 128 R-sq: within = 0.8277 Obs per group: min = 4 between = 0.8305 avg = 5.9 overall = 0.8281 max = 6 Random effects u_i ~ Gaussian Wald chi2(3) = 3633.42 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 rderj Coef. Std. Err. z P> z [95% Conf. Interval] rexs -1.234984 53.5979-0.02 0.982-106.2849 103.815 rdfras -62.30774 1.929159-32.30 0.000-66.08882-58.52665 rdfrli -13.35797.646535-20.66 0.000-14.62516-12.09078 _cons 26.26059 15.22163 1.73 0.084-3.57326 56.09444 sigma_u 0 sigma_e 383.88777 rho 0 (fraction of variance due to u_i). estimates store random. hausman fixed random Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rexs 29.5331-1.234984 30.76809 85.76539 rdfras -62.12105-62.30774.1866859.8862267 rdfrli -13.37106-13.35797 -.0130911.2974712 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(3) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 0.19 Prob>chi2 = 0.9787 39

Fixed-effects (within) regression Number of obs = 756 Group variable: id Number of groups = 126 R-sq: within = 0.4553 Obs per group: min = 6 between = 0.3272 avg = 6.0 overall = 0.4110 max = 6 F(6,624) = 86.92 corr(u_i, Xb) = -0.1280 Prob > F = 0.0000 req Coef. Std. Err. t P> t [95% Conf. Interval] rnfadkurs L1. -88.5532 4.976725-17.79 0.000-98.32636-78.78005 rnfa L1. 12.6412.9196849 13.75 0.000 10.83515 14.44726 dkurs L1. -14.13346 2.937476-4.81 0.000-19.902-8.364924 netkumderd~s L1. -.020979.0656822-0.32 0.750 -.1499639.1080059 dli L1..0516997.008097 6.39 0.000.0357991.0676003 rwc.1373136.2018649 0.68 0.497 -.2591032.5337304 _cons -2.269386 2.842485-0.80 0.425-7.851381 3.312608 sigma_u 2.9689161 sigma_e 4.4181256 rho.31108811 (fraction of variance due to u_i) F test that all u_i=0: F(125, 624) = 2.01 Prob > F = 0.0000 40

. xtreg rni rdert rexs rfrli rfras rli rsa, fe Fixed-effects (within) regression Number of obs = 761 Group variable: id Number of groups = 128 R-sq: within = 0.5288 Obs per group: min = 4 between = 0.1373 avg = 5.9 overall = 0.3041 max = 6 F(6,627) = 117.29 corr(u_i, Xb) = -0.6040 Prob > F = 0.0000 rni Coef. Std. Err. t P> t [95% Conf. Interval] rdert 3.19e-06.0000114 0.28 0.780 -.0000193.0000256 rexs -.0080938.0703776-0.12 0.908 -.1462982.1301106 rfrli.9951559.0427459 23.28 0.000.9112135 1.079098 rfras -.5369794.1547834-3.47 0.001 -.8409361 -.2330227 rlia -1.312004.0545632-24.05 0.000-1.419153-1.204856 rsa.0391975.0263271 1.49 0.137 -.0125024.0908975 _cons.5562386.0475632 11.69 0.000.4628362.6496411 sigma_u.28640161 sigma_e.26771745 rho.53368038 (fraction of variance due to u_i) F test that all u_i=0: F(127, 627) = 3.74 Prob > F = 0.0000. estimates store fixed. xtreg rni rdert rexs rfrli rfras rli rsa, re Random-effects GLS regression Number of obs = 761 Group variable: id Number of groups = 128 R-sq: within = 0.5074 Obs per group: min = 4 between = 0.1738 avg = 5.9 overall = 0.3429 max = 6 Random effects u_i ~ Gaussian Wald chi2(6) = 525.56 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 rni Coef. Std. Err. z P> z [95% Conf. Interval] rdert 4.00e-06.0000118 0.34 0.735 -.0000192.0000272 rexs -.0036202.0530221-0.07 0.946 -.1075416.1003013 rfrli.8390633.040695 20.62 0.000.7593025.9188241 rfras -.6446199.1142258-5.64 0.000 -.8684984 -.4207414 rlia -.9202645.0433631-21.22 0.000-1.005255 -.8352743 rsa.0553108.0102943 5.37 0.000.0351343.0754872 _cons.3567872.0296995 12.01 0.000.2985773.414997 sigma_u.12863863 sigma_e.26771745 rho.18757422 (fraction of variance due to u_i). estimates store random. hausman fixed random Note: the rank of the differenced variance matrix (5) does not equal the number of coefficients being tested (6); be sure this is what you expect, or there may be problems computing the test. Examine the output of your estimators for anything unexpected and possibly consider scaling your variables so that the coefficients are on a similar scale. Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rdert 3.19e-06 4.00e-06-8.01e-07. rexs -.0080938 -.0036202 -.0044736.0462781 rfrli.9951559.8390633.1560926.0130814 rfras -.5369794 -.6446199.1076405.1044527 rlia -1.312004 -.9202645 -.3917398.0331177 rsa.0391975.0553108 -.0161132.024231 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(5) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 203.37 Prob>chi2 = 0.0000 (V_b-V_B is not positive definite) 41

. xtreg rebit rdert rexs rfrli rfras rli rsa, fe Fixed-effects (within) regression Number of obs = 761 Group variable: id Number of groups = 128 R-sq: within = 0.3886 Obs per group: min = 4 between = 0.0875 avg = 5.9 overall = 0.1850 max = 6 F(6,627) = 66.41 corr(u_i, Xb) = -0.6087 Prob > F = 0.0000 rebit Coef. Std. Err. t P> t [95% Conf. Interval] rdert 5.42e-07.0000253 0.02 0.983 -.0000491.0000502 rexs -.0065769.1557404-0.04 0.966 -.3124128.299259 rfrli 1.260503.0945934 13.33 0.000 1.074745 1.446261 rfras -.674735.3425241-1.97 0.049-1.347368 -.0021017 rlia -2.334061.1207443-19.33 0.000-2.571174-2.096949 rsa.205332.0582599 3.52 0.000.0909239.3197401 _cons.8601732.1052538 8.17 0.000.6534806 1.066866 sigma_u.65253708 sigma_e.59243856 rho.54816062 (fraction of variance due to u_i) F test that all u_i=0: F(127, 627) = 4.24 Prob > F = 0.0000. estimates store fixed. xtreg rebit rdert rexs rfrli rfras rli rsa, re Random-effects GLS regression Number of obs = 761 Group variable: id Number of groups = 128 R-sq: within = 0.3742 Obs per group: min = 4 between = 0.0979 avg = 5.9 overall = 0.2043 max = 6 Random effects u_i ~ Gaussian Wald chi2(6) = 299.53 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 rebit Coef. Std. Err. z P> z [95% Conf. Interval] rdert 4.88e-06.0000258 0.19 0.850 -.0000457.0000555 rexs.0442035.1234381 0.36 0.720 -.1977307.2861377 rfrli 1.124642.0904729 12.43 0.000.9473188 1.301966 rfras -.7627092.2666267-2.86 0.004-1.285288 -.2401305 rlia -1.702594.0996648-17.08 0.000-1.897933-1.507254 rsa.1090677.025429 4.29 0.000.0592277.1589077 _cons.6499244.0722979 8.99 0.000.5082232.7916256 sigma_u.36824464 sigma_e.59243856 rho.27868389 (fraction of variance due to u_i). estimates store random. hausman fixed random Note: the rank of the differenced variance matrix (5) does not equal the number of coefficients being tested (6); be sure this is what you expect, or there may be problems computing the test. Examine the output of your estimators for anything unexpected and possibly consider scaling your variables so that the coefficients are on a similar scale. Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rdert 5.42e-07 4.88e-06-4.34e-06. rexs -.0065769.0442035 -.0507805.0949637 rfrli 1.260503 1.124642.1358605.0276145 rfras -.674735 -.7627092.0879742.2150185 rlia -2.334061-1.702594 -.6314678.0681623 rsa.205332.1090677.0962643.0524173 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(5) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 99.99 Prob>chi2 = 0.0000 (V_b-V_B is not positive definite) 42

. xtreg ropx rdert rexs rfrli rfras rli rsa, fe Fixed-effects (within) regression Number of obs = 761 Group variable: id Number of groups = 128 R-sq: within = 0.4384 Obs per group: min = 4 between = 0.0235 avg = 5.9 overall = 0.1447 max = 6 F(6,627) = 81.58 corr(u_i, Xb) = -0.6792 Prob > F = 0.0000 ropx Coef. Std. Err. t P> t [95% Conf. Interval] rdert -1.16e-07.0000213-0.01 0.996 -.000042.0000417 rexs.0465644.1311983 0.35 0.723 -.2110769.3042056 rfrli -1.009748.079687-12.67 0.000-1.166234 -.853262 rfras.5634351.288548 1.95 0.051 -.0032024 1.130073 rlia 2.214216.101717 21.77 0.000 2.014469 2.413964 rsa -.131323.0490791-2.68 0.008 -.2277023 -.0349437 _cons -.6788994.0886675-7.66 0.000 -.8530207 -.5047782 sigma_u.66185087 sigma_e.49908008 rho.63750406 (fraction of variance due to u_i) F test that all u_i=0: F(127, 627) = 5.15 Prob > F = 0.0000. estimates store fixed. xtreg ropx rdert rexs rfrli rfras rli rsa, re Random-effects GLS regression Number of obs = 761 Group variable: id Number of groups = 128 R-sq: within = 0.4157 Obs per group: min = 4 between = 0.0406 avg = 5.9 overall = 0.1700 max = 6 Random effects u_i ~ Gaussian Wald chi2(6) = 292.87 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000 ropx Coef. Std. Err. z P> z [95% Conf. Interval] rdert -2.76e-06.0000226-0.12 0.903 -.0000471.0000416 rexs -.1241271.1083961-1.15 0.252 -.3365796.0883253 rfrli -.8687117.0793794-10.94 0.000-1.024293 -.7131309 rfras.4832587.2341484 2.06 0.039.0243362.9421812 rlia 1.467285.0874996 16.77 0.000 1.295789 1.638781 rsa -.1085599.0223581-4.86 0.000 -.1523809 -.0647388 _cons -.2823382.0635496-4.44 0.000 -.4068931 -.1577833 sigma_u.31167679 sigma_e.49908008 rho.28057732 (fraction of variance due to u_i). estimates store random. hausman fixed random Note: the rank of the differenced variance matrix (5) does not equal the number of coefficients being tested (6); be sure this is what you expect, or there may be problems computing the test. Examine the output of your estimators for anything unexpected and possibly consider scaling your variables so that the coefficients are on a similar scale. Coefficients (b) (B) (b-b) sqrt(diag(v_b-v_b)) fixed random Difference S.E. rdert -1.16e-07-2.76e-06 2.64e-06. rexs.0465644 -.1241271.1706915.073914 rfrli -1.009748 -.8687117 -.1410361.006995 rfras.5634351.4832587.0801764.1686252 rlia 2.214216 1.467285.7469313.0518668 rsa -.131323 -.1085599 -.0227631.0436906 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(5) = (b-b)'[(v_b-v_b)^(-1)](b-b) = 186.64 Prob>chi2 = 0.0000 (V_b-V_B is not positive definite) 43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61