I am trying to run a panel unit root test using the PESCADF command in Stata. However I am confused about selecting the lag order. I find that when I keep my lag order at 4, the test concludes that that the variable is stationary, and if if I bring it any lower the test results shows that variable is non-stationary. Could anyone please advise me on what to do? Is there a criteria for selecting lag order? Thanks in advance.
Related Posts with Selecting lags for Pesarans unit root tests which controls for cross-sectional dependence
Calling a block of code within a DO fileHello All, I am running a DO file to create a table of summary statistics for a large number of var…
looping over many regressionsHello I am trying to run 11 separate regressions. One with each of the local vars as a single predic…
Regression help on statistical significanceHi all! I am an econ student finishing my bachelor thesis and I have an econometrics section coming…
xtprobit random effects negative varianceDear all stata users, I have several variables that explains the state (approved (aprov)=1; not app…
How to change local stringsLet's say I want to change the content of string local named asdf from q to 2. I tried this Code: …
Subscribe to:
Post Comments (Atom)
0 Response to Selecting lags for Pesarans unit root tests which controls for cross-sectional dependence
Post a Comment