PHD: stata

Predict yhat
Predict residual

LATE

An F statistic is a value you get when you run an ANOVA test or a regression analysis to find out if the means between two populations are significantly different. It’s similar to a T statistic from a T-Test; A-T test will tell you if a single variable is statistically significant and an F test will tell you if a group of variables are jointly significant.

General introductory stats:

http://courses.umass.edu/pubp608/

Chi-squared test:

https://www.youtube.com/watch?v=WXPBoFDqNVk

Contingency table:

https://us.sagepub.com/sites/default/files/upm-binaries/43350_4.pdf

Asses the elasticity of substitution

Assess the lq

Assess the elasticity equation in general

Need Reconsideration ...

Problem of Endogeneity

http://www.homepages.ucl.ac.uk/~uctpsc0/Teaching/GR03/Endogeneity.pdf

http://www.slideshare.net/KelleyMacEwen/endogeneity-exogeneity

http://www.sfu.ca/~pendakur/teaching/buec333/Multicollinearity%20and%20Endogeneity.pdf

In general, we say that a variable X is endogenous if it is correlated with the model error term. Endogeneity always induces bias.

http://www.soc.aau.dk/fileadmin/user_upload/kbm/VoF/Kurser/2011/kvantitative-metoder/Slides/TR-Endogeneity.pdf

What is GLS?

GLS --> compress residuals that are far apart and pull close ones.

What is poisson dist and negative binomial?

· Wooldrige test for autocorrelation

· Control for population separately. Run the xtreg.

· Control for year in the pooled OLS. (What is the good of pooled ols or xtreg in panel analysis?)

Notes from Min’s presentation

Page 60-61 Graph.

We use a t-distribution on regressions

You want y-hat minus y-bar to be minimal (ESS). This is good. Close to mean. Small S.D.

y minus y-hat big is not good. (RSS)

* Log-Linear % to Unit

reg lncigsale retprice beer lnincome age15to24

* Linear-Log Unit to %

reg cigsale lnretprice beer lnincome age15to24

* Log-Log (Elasticity) % to %

reg lncigsale lnretprice beer lnincome age15to24

Run these tests after regression

** Breusch-Pagan Test

qui reg cigsale retprice beer lnincome age15to24

estat hettest

** White test

qui reg cigsale retprice beer lnincome age15to24

estat imtest, white

Robust Regression:

multicol

distributional

heteroscedasticity (White, Brusch-Pagan)

Outliers (Cook’s Distance)

PCA

https://www.princeton.edu/~otorres/Factor.pdf

http://core.ecu.edu/psyc/wuenschk/MV/SEM/Path.pdf

http://www.unt.edu/rss/class/mike/6810/Path.pdf

Data Reduction process

https://www.youtube.com/watch?v=MB-5WB3eZI8

FE RE

Video Tutorial: https://www.youtube.com/watch?v=pQvhi60yN74

Material: http://forscenter.ch/wp-content/uploads/2013/11/Lipps_Kuhn_SummerSchool_Lugano2012_slides.pdf

Analytics:

Network:

https://sites.google.com/site/ucinetsoftware/