Hi all,
I attach a mail that I received from Johannes Pfeifer, which raises 3 problems and makes 2 suggestions.
I pushed a fix for problem #2 (varlist_indices.m) and I can’t reproduce problem #3 (related to USE_DLL) even under Windows.
Concerning problem #1 (empirical autocorrelations), Johannes is right in the sense that - with the current code - one can get an autocorrelation that is not within [-1,1], because the covariance on the numerator does not span the same sample than the variances on the denominator; the fix that he suggests would fix this. On the other hand, estimators of the autocorrelation coefficients are known to be biased in finite samples (and there is no easy way of fixing this), so applying Johannes’ change would just consist in moving from a biased estimator to another biased estimator (and I don’t know how the bias is changed). These estimators are only asymptotically consistent, and getting a weird value is the indication that the sample is too small and that more simulations should be done.
What is your opinion about this issue, and about the two suggestions (options for no IRF files, and NaN/Inf treatment in estimation) ?
Best,
On 01/10/2011 05:25 PM, Sébastien Villemot wrote:
Hi all,
I attach a mail that I received from Johannes Pfeifer, which raises 3 problems and makes 2 suggestions.
I pushed a fix for problem #2 (varlist_indices.m) and I can’t reproduce problem #3 (related to USE_DLL) even under Windows.
On devrait peut-etre utiliser clear <modname>_dynamic <modname>_static
a la fin du *.mod file
Concerning problem #1 (empirical autocorrelations), Johannes is right in the sense that - with the current code - one can get an autocorrelation that is not within [-1,1], because the covariance on the numerator does not span the same sample than the variances on the denominator; the fix that he suggests would fix this. On the other hand, estimators of the autocorrelation coefficients are known to be biased in finite samples (and there is no easy way of fixing this), so applying Johannes’ change would just consist in moving from a biased estimator to another biased estimator (and I don’t know how the bias is changed). These estimators are only asymptotically consistent, and getting a weird value is the indication that the sample is too small and that more simulations should be done.
What is your opinion about this issue, and about the two suggestions (options for no IRF files, and NaN/Inf treatment in estimation) ?
Let's compute the covariance on the minimal sample as suggested by Johannes
Best
Michel
Best,
Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
Michel Juillard michel.juillard@mjui.fr writes:
On 01/10/2011 05:25 PM, Sébastien Villemot wrote:
Concerning problem #1 (empirical autocorrelations), Johannes is right in the sense that - with the current code - one can get an autocorrelation that is not within [-1,1], because the covariance on the numerator does not span the same sample than the variances on the denominator; the fix that he suggests would fix this. On the other hand, estimators of the autocorrelation coefficients are known to be biased in finite samples (and there is no easy way of fixing this), so applying Johannes’ change would just consist in moving from a biased estimator to another biased estimator (and I don’t know how the bias is changed). These estimators are only asymptotically consistent, and getting a weird value is the indication that the sample is too small and that more simulations should be done. What is your opinion about this issue, and about the two suggestions (options for no IRF files, and NaN/Inf treatment in estimation) ?Let's compute the covariance on the minimal sample as suggested by Johannes
Ok, I merged the change he suggested.
-
Michel Juillard -
Sébastien Villemot