Ridge Regression for Panel Data
Hi, I am doing a unit-fixed effects panel data analysis to examine the relationship between income and education for 26 regions over 9 years. My dependent variable is Gini, independent variables are proportion of graduates from various levels of education to total population and my controls are regional GDP, regional CPI, regional unemployment, regional LFPR and net migration rate. However, my model suffers from multicollinearity. I've been looking at regularisation techniques and want to use Ridge but I am not sure if it is suitable for panel data. Can anyone give me advice/recommendations on this please? Thanks very much in advance!!
Skalonga
7
Join Matchmaticians Affiliate Marketing
Program to earn up to a 50% commission on every question that your affiliated users ask or answer.
- unanswered
- 193 views
- Pro Bono
Related Questions
- ANCOVA: R Squared and Partial Eta Squared
- What does it mean for regression to only be normally distributed in the vertical direction and correlation to be normally distributed in horizontal and vertical directions?
- Bayesian statistics
- applied probability
- Bernoulli problem
- The derivation of the formula for variance for a Pareto Distribution
- How do I compare categorical data with multiple uneven populations?
- CLT and probability
Questions at this level should come with a bounty.
How much do you reckon?
Think about how much time one may need to spent to write an answer and what a reasonable hourly rate for such individual would be. It'll give you a pretty good idea.