In addition to the manual entry, note that the code is accessible and
fairly short. So
. viewsource pergram.ado
followed by detailed scrutiny may answer your question.
Nick
[email protected]
Dan Weitzenfeld
I'm wondering if anyone can confirm that I am correctly interpreting
the results of the command -pergram- with the option [,
generate(newvar)].
As per the digital signal processing book I have, if you have _N
samples, your DFT results are binned into _N bins, with:
bin _n
holding the results for the frequency
(_n*samplingfrequency)/_N
So does Stata use this methodology for putting the "raw periodogram
values" into newvar?
Am I correct that because Stata does not have an observation zero, it
puts the results for the frequency (_n-1*samplingfrequency)/_N in
observation _n , with _n ranging from observation f to l?
Any guidance here would be much appreciated; Stata's help for this
command is anemic.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
