Scribd
Upload
43 views13 pages

Analysing Panel Data Using STATA

This document provides a step-by-step guide for analyzing panel data using STATA 16.0, including importing data, declaring it as panel data, and performing various statistical tests and regressions. It covers commands for descriptive statistics, correlations, OLS regression, robustness tests, and panel regression techniques such as Hausman and LM tests. Additionally, it explains how to export results from STATA to MS Word or Excel.

Uploaded by

Amos Ganyam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
43 views13 pages

Analysing Panel Data Using STATA

This document provides a step-by-step guide for analyzing panel data using STATA 16.0, including importing data, declaring it as panel data, and performing various statistical tests and regressions. It covers commands for descriptive statistics, correlations, OLS regression, robustness tests, and panel regression techniques such as Hausman and LM tests. Additionally, it explains how to export results from STATA to MS Word or Excel.

Uploaded by

Amos Ganyam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Analysing Panel Data

using STATA 16.0


By
Amos Ganyam
Step 1: Import Data to STATA
• Open STATA from Desktop, Start Menu or Task Bar
by clicking on the icon.
• Import Data from MS Excel either by copying and
pasting into the Data editor or by navigating
through the file menu.
Step 2: Declare Data to be Panel
• Convert your grouping variable from strings to
numbers if it was not done manually in Excel.
• Type this command
egen newID = group(oldID)

• Declare data as panel using the follow commands


xtset newID
xtset newID year, yearly
Step 3: Descriptive Statistics
• Type this command
summarize y x1 x2…xn, separator(0)
• Separator divides the table into sections depending
on specification
Step 4: Correlations
• Type this command
pwcorr y x1 x2…xn, star(5)
• Star (5) shows an asterisk for significant
relationships
Step 5: OLS regression
• Type this command
regress y x1 x2…xn
• List all the variables after regress starting with the
dependent variable
Step 6: Robustness Tests-Normality
• After the OLS command, type this command
predict residual,r
• This saves the residual as a variable for normality
estimation.
• For Histogram normality test, type this command:
histogram residual,normal
• For Jarque-Bera normality test, type this command:
jb residual
• For Shapiro Wilk test, type this command:
swilk residual
Step 6 Continued
• For Skewness and Kurtosis test, type this command:
sktest residual
• For VIF, type this command:
vif
• For Heteroscedasticity, type this command:
hettest
Step 7: Panel Regression-Hausman Test
• First perform your fixed effects with this command:
xtreg y x1 x2 x3…xn,fe
• Store the fixed effects results with this command:
estimates store fixed
• Then perform the random effects with this command:
xtreg y x1 x2 x3…xn,re
• Store the random effects results with this command:
estimates store random
• Now perform Hausman test with the stored results by
typing this command:
hausman fixed random
Step 8: Panel Regression-LM Test
• Perform the random effects again with this command:
xtreg y x1 x2 x3….xn,re
• Now perform LM test by typing this command:
xttest0
Step 9: Regression with Robust
Standard Errors
• For random effects, use this command:
xtreg y x1 x2 x3…xn,re robust
• For fixed effects, use this command:
xtreg y x1 x2 x3…xn,fe robust
• For pooled OLS, use this command:
regress y x1 x2 x3…xn,robust
Exporting Results from STATA
• Copying and pasting results from STATA to Ms Word
or Ms Excel
Highlight the desired result, select copy as picture, go to
Ms Word desired document and paste.
Highlight the desired result, select copy as HTML, go to
Ms Word desired document and paste for easy
extraction of desired components
Thanks for
listening.

You might also like