standard deviation xtpoisson

[Federico Nutarelli]

Hi all,

is there a way to recover standard deviations of coefficients from xtpoisson, fe?
in particular, I have:

quietly xtpoisson trials y residuals average_age_prodbyatc3 avg_prd_sq mean_agefirm_byatc mean_agefirm_squared hhi share_expired share_patented i.Year, fe vce(robust)

and would like to recover and save the standard deviation of coefficients. Have I provided too few info? Please let me know

Thanks for the help in advance

[Clyde Schechter]

A coefficient is a constant. It does not have a standard deviation. There is no such thing. Perhaps you mean the standard error? If so you will find them in _se. For example you can save these in local macros or create variables that contain them with code like:

local se_of_y = _se[y]

gen y_std_err = _se[y]

[Carlo Lazzaro]

Federico:
as an aside to Clyde's helpful reply, standard error is, after all. the standard deviation of the sample distribution of a given parameter (coefficient, a far as regression is concerned).
Are you asking for the standard deviation of the variables that enter the regression model as predictors?

[Federico Nutarelli]

Thank you Carlo and Clyde for the kind reply.
To clarify: I know that standard deviation is just standard error multiplied the square root of the sample size. Hence since I am performing bootstrap on different sample sizes (i.e. I am proceeding by trial and error to find the correct size for the bootstrap sample), I was just asking myself if stata provides a command to directly find standard deviation. By the way, I thing standard error should be fine.

Further, since you are both tenured members, I would like to ask if you can, please, give me some advices on how to better post. Indeed my last post on the bootstrap procedure, which is still my major concern did not receive any attention. I do not want to repost it many times since it is not much respectful for people on the forum but since I provided data and an explanation of what I would like to perform I cannot understand the reason why it went quite unseen.

[Carlo Lazzaro]

Federico:
see my today's reply to your https://www.statalist.org/forums/for...rap-two-stages