How to compute simulation error of a parameter in a simulated data set

Accepted Solution Solved
Reply
Super Contributor
Posts: 297
Accepted Solution

How to compute simulation error of a parameter in a simulated data set

I am interested in computing the simulated error for the parameter t in the regression.  SAS gives me the regression error ot t which is StdEr I wish to know if sas has an easy way to get it.

THanks


Accepted Solutions
Solution
‎02-27-2014 08:45 AM
Respected Advisor
Posts: 2,655

Re: How to compute simulation error of a parameter in a simulated data set

Would restating the simulation error as sqrt(sum(beta_i - beta_bar)/(N-1)) be correct?

I would feel more comfortable with sqrt(sum((beta_i - beta_bar)^2)/(N-1)), so that it is now essentially the standard deviation of the beta_i estimates.

Steve Denham

Message was edited by: Steve Denham

View solution in original post


All Replies
Super User
Posts: 17,963

Re: How to compute simulation error of a parameter in a simulated data set

Is your formula above correct?

desireatem wrote:

The simulated error is sqrt (1/N-1(SUM(bete_i-bar(beta)).

Super Contributor
Posts: 297

Re: How to compute simulation error of a parameter in a simulated data set

It might not be correctly type. However do you know how to compute the simulated standard error. This is the correct simulation error: SQRT [1/N-1 *SUM(Est(beta_i- bar(beta))] where N=1000 number of replicates.

Solution
‎02-27-2014 08:45 AM
Respected Advisor
Posts: 2,655

Re: How to compute simulation error of a parameter in a simulated data set

Would restating the simulation error as sqrt(sum(beta_i - beta_bar)/(N-1)) be correct?

I would feel more comfortable with sqrt(sum((beta_i - beta_bar)^2)/(N-1)), so that it is now essentially the standard deviation of the beta_i estimates.

Steve Denham

Message was edited by: Steve Denham

Super User
Posts: 17,963

Re: How to compute simulation error of a parameter in a simulated data set

If I'm interpreting this correctly,

SUM(bete_i-bar(beta))=0 always, assuming bar(beta) is average of beta.

🔒 This topic is solved and locked.

Need further help from the community? Please ask a new question.

Discussion stats
  • 4 replies
  • 832 views
  • 0 likes
  • 3 in conversation