Reducing Bootstrap Panel Data Regression Time

[Louis Kos]

Hello everyone,

I am currently in the process of running a panel data regression and wish to obtain bootstrapped p-values. The code is shown below:

 xtset id date
xtreg ret log_age iporet_nyse ipovol oil bank, vce(bootstrap, reps(1000) seed(1234)) cluster(id)

It does function, however, my dataset contains 1.2m observations (with only relevant variables included), and as such the regression will take 9 hours for 1000 reps. I would like to be able to do 1000 reps. Is there any way to hasten the process, or obtain the bootstrapped p-values in another way?

Many thanks in advance.

[Carlo Lazzaro]

Louis:
in An Introduction to the Bootstrap. New York: Chapman & Hall/CRC, 1993 (page 47) Efron, B., and R. J. Tibshirani state that 25-200 bootstrap replications are enough for standard errors (in most researches you'll find 200).