I'm using the autocorrelation command ac and the partial autocorrelation command pac. I'm running them on over 1,000 different time series with length T=12. May I ask a couple of questions?
The commands are slow because they generate a plot, which I don't want to see. Is there a way to save time by suppressing the plot? It's no big deal if you have only one time series, but if you have over 1,000, it's a drag.
The pac command won't calculate the partial autocorrelation for lags beyond 4. Why not? Is this an arbitrary software limit, or is there some mathematical reason you can't calculate the lag-5 autocorrelation from a time series with T=12.
Many thanks.
Related Posts with Autocorrelation (ac and pac)
Matching based on SIZE and Industry . dataex ID TIC SIZE PROT Industry_code LN_AUDFEE ----------------------- copy starting from the ne…
How do I refer to a variable that might not be included in the varlist of a program?How do I refer to a variable that might not be included in the varlist of a program? I have simple …
Specifying a Matrix in the bayesmh command.Good evening! I'm trying to fit a Poisson Gamma Mixture in Stata. One of the parameters of interest…
Repeating 1 annual data point 4 timesHi, how to I repeat 1 annual datapoint in order to give me 4 quarterly data points to match the rest…
Crosstab of two variables, specifying a single value for each variableHello Statalist, I'm sure this is a very simply question for most of you. I have two variables, Q4…
Subscribe to:
Post Comments (Atom)
0 Response to Autocorrelation (ac and pac)
Post a Comment