2
$\begingroup$

I performed multiple pooled cross-sectional regressions with the same time-intervals (5years) in different years. I'm wondering now on how to aggregate the different regression outputs. Does it make sense to average the estimated coefficients or do you rather use one simple pooled regression over all $t$s? And assuming taking averages makes sense, how can you get a valid $t$-statistics?

Here is an extract of the code I used; thank you for your help. 

> head(Grunfeld)
  firm year   inv  value capital
1    1 1935 317.6 3078.5     2.8
2    1 1936 391.8 4661.7    52.6
3    1 1937 410.6 5387.1   156.9
4    1 1938 257.7 2792.2   209.2
5    1 1939 330.8 4313.2   203.4
6    1 1940 461.2 4643.9   207.2

library(plm)
data("Grunfeld", package="plm")

# Store each subset regression in myregression
myregression <- list()

# Regression on six-year subsets of Grunfeld
for(t in 1940:1950) {

  myregression[[t-1939]] <- lm(inv ~ value + capital, 
                              subset(Grunfeld, year<=t & year>=t-5))

  # Rename list elements by year range of subset
  names(myregression)[[t-1939]] = paste0("Years",t)
}
Improve this question
$\endgroup$
2
  • 2
    $\begingroup$ My suggestion would be to put whatever is necessary to understand the question directly in your question, instead of linking to a post elsewhere. In other words, it'd be best if your question was standalone. $\endgroup$
    – Patrick Coulombe
    Commented Jul 13, 2014 at 23:16
  • 1
    $\begingroup$ No one came up with a similar problem? or is it that obvious? please help $\endgroup$
    – Gritti
    Commented Jul 14, 2014 at 19:40

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.