EqBootstrap (bootstrap standard errors)
Moderators: EViews Gareth, EViews Moderator, EViews Esther

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Re: EqBootstrap (bootstrap standard errors)
No addins work with the student version. No, this addin does not bootstrap impulse responses.
Follow us on Twitter @IHSEViews
Re: EqBootstrap (bootstrap standard errors)
I am trying to install the AddIn in Eviews but it is not working.
What is the procedure for this?
What is the procedure for this?

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Re: EqBootstrap (bootstrap standard errors)
Same with all addins. Select it from the list, and hit the install button.
Follow us on Twitter @IHSEViews
Re: EqBootstrap (bootstrap standard errors)
Hi,
 Is it appropriate to use this on timeseries data?
 Is there a way to save the bootstrapped standard errors in a vector so that I can then use them in later calculations (without going into the results table and manually copying the bootstrapped standard errors into a vector myself)?
(I am using Eviews 7.2)
Thanks in advance for your assistance!
 Is it appropriate to use this on timeseries data?
 Is there a way to save the bootstrapped standard errors in a vector so that I can then use them in later calculations (without going into the results table and manually copying the bootstrapped standard errors into a vector myself)?
(I am using Eviews 7.2)
Thanks in advance for your assistance!

 Posts: 47
 Joined: Sat Oct 11, 2008 9:37 am
 Location: Oslo
 Contact:
Re: EqBootstrap (bootstrap standard errors)
How do I get access to the bootstrapresults? Since it's not stored, using eg. equation.@tstats(2) won't work.
I tried the following, and it didn't work.
It runs the bootstrap and show me the results in the equation window. But seems I have to manually freeze the output.
Thanks in advance,
Dagfinn
I tried the following, and it didn't work.
Code: Select all
freeze(mode=overwrite,_temptab2) _tmpeq.eqbootstrap(bs=5000,se,res)
It runs the bootstrap and show me the results in the equation window. But seems I have to manually freeze the output.
Thanks in advance,
Dagfinn

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Re: EqBootstrap (bootstrap standard errors)
You'll have to modify the addin to store the table of results somewhere. Actually, there's probably a line at the bottom of the addin that deletes the output table after displaying it in the current equation. Just remove that line.
Follow us on Twitter @IHSEViews
Re: EqBootstrap (bootstrap standard errors)
Hello,
I have faced a problem with bootstrapping. There are a number of outcomes one at a time when I try to do Bootstrapping using Addins > Bootstrap Coefficients. One outcome is that I receive an error saying "Procedure can only be run from equation estimated by list".
But sometimes I can open the bootstrap window and after clicking on Bootstrap Coefficients, and choosing the "Resample Residual" option and setting the number of repetitions to say (10000), I will be prompted to the estimate equation window with this specification (_yres01 fx ipi market c), and it keeps estimating anytime i press the OK button.
I remember there was another outcome as well, which after choosing the "Resample Residual" and setting number of repetitions, the analysis was performed only at one time, where there was only one estimation output.
I would be very grateful if you could help me out how I can deal with this problem and what each of these three outcomes mean. I am using Eviews 8. Thank you
I have faced a problem with bootstrapping. There are a number of outcomes one at a time when I try to do Bootstrapping using Addins > Bootstrap Coefficients. One outcome is that I receive an error saying "Procedure can only be run from equation estimated by list".
But sometimes I can open the bootstrap window and after clicking on Bootstrap Coefficients, and choosing the "Resample Residual" option and setting the number of repetitions to say (10000), I will be prompted to the estimate equation window with this specification (_yres01 fx ipi market c), and it keeps estimating anytime i press the OK button.
I remember there was another outcome as well, which after choosing the "Resample Residual" and setting number of repetitions, the analysis was performed only at one time, where there was only one estimation output.
I would be very grateful if you could help me out how I can deal with this problem and what each of these three outcomes mean. I am using Eviews 8. Thank you

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Re: EqBootstrap (bootstrap standard errors)
The first issue is caused by the fact that the addin only supports equations estimated by list. If you've estimated an equation by expression, it won't work.
We'd have to see examples of the other two.
We'd have to see examples of the other two.
Follow us on Twitter @IHSEViews
Re: EqBootstrap (bootstrap standard errors)
The first issue is still an issue for me. Im sorry but I need a clarification. Estimating the equation by list is simply highlighting the dependent followed by independent variables, and then right click > open > as equation, right? I actually did so and yet i receive the same error.

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Re: EqBootstrap (bootstrap standard errors)
We'd have to see it then.
Follow us on Twitter @IHSEViews

 Posts: 2
 Joined: Fri Oct 31, 2014 9:56 am
Re: EqBootstrap (bootstrap standard errors)
EViews Gareth wrote:You'll have to modify the addin to store the table of results somewhere. Actually, there's probably a line at the bottom of the addin that deletes the output table after displaying it in the current equation. Just remove that line.
Could you please be more specific on that. I have the same problem: I need the results of the coefficients and the corresponding standard errors. I was able to change the addin, so that i get a vector of all bootstrapped coefficients. But how can I get the corresponding standard errors in a vector?
Thanks in advance!

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Re: EqBootstrap (bootstrap standard errors)
They're stored in the matrix {%cov}. That matrix gets deleted at the end of the bootstrap.prg file. Just remove the entry that deletes it.
Follow us on Twitter @IHSEViews

 Posts: 2
 Joined: Fri Oct 31, 2014 9:56 am
Re: EqBootstrap (bootstrap standard errors)
Thanks for your answer!
But still there is an open question:
If i delete {%cov} at the end, I get the covariance matrix of the centered equation of all bootstrapped coefficients. So just the variance for one coefficient.
Is it anyhow possible to get all standard errors of all 10.000 bootstrapped coefficients? I have the 10.000 bootstrapped coefficients in a vector and i want to calculate the distribution of all the tvalues of these 10.000 coeffcients.
But still there is an open question:
If i delete {%cov} at the end, I get the covariance matrix of the centered equation of all bootstrapped coefficients. So just the variance for one coefficient.
Is it anyhow possible to get all standard errors of all 10.000 bootstrapped coefficients? I have the 10.000 bootstrapped coefficients in a vector and i want to calculate the distribution of all the tvalues of these 10.000 coeffcients.

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12501
 Joined: Tue Sep 16, 2008 5:38 pm
Who is online
Users browsing this forum: No registered users and 9 guests