EViews.com

Hello,

I am trying to apply the HAC in OLS estimation. I noticed that once we specified prewithitening, the standard error of some coefficients may turn to NA. Can you please help? Please see the code I used. Thanks.

eq_loop.ls(COV=HAC, COVLAG=1, COVKERN=QUADSPEC, COVBW=ANDREWS) {%depend} c {%regressor}

Are there any warning messages produced on the equation output?

Thanks for your help.

There is no error message. But I want to get the standard error value, which is NA in the model output.

Method: Least Squares
Date: 06/30/16 Time: 11:51
Sample (adjusted): 2005Q4 2013Q4
Included observations: 33 after adjustments
HAC standard errors & covariance (Prewhitening with lags = 1,
Quadratic-Spectral kernel, Andrews bandwidth = 0.6120)

Variable Coefficient Std. Error t-Statistic Prob.

C -171.7701 NA NA NA

Post your workfile.

Sorry, I cannot directly post the workfile on the website. Could I email you my workfile?

Thanks.

Please make sure your copy of EViews is up to date. Your estimation worked fine for us.