The Intuition behind Panel Regression

Data, R code, and slides available here:
https://github.com/alexcoad/Econometrics

This video introduces panel regression, and explains why it differs from standard OLS linear regression. Using an example of size of police force and murders across a panel of 5 cities, the video shows how estimation using a standard OLS linear regression gets misleading results. Panel regression techniques, in contrast, can obtain more sensible results, by taking into account unobserved time-invariant individual-specific heterogeneity (in the form of eta_i). The video discusses types of data structure, and explains why fixed effects regression is similar to a regression where each individual has their own individual-specific intercept.

The video discusses various panel data regression models: fixed effects, differences, and the LSDV (least squares dummy variable) estimator, discussing also the differences between these three panel data techniques.

The video shows figures and graphs as well as some code in R software, although viewers can follow the video without any knowledge of R software.
The R code and slides that accompany this video are freely available from my github page:

https://github.com/alexcoad/Econometrics