The STEPS instruction crashes in RATS 9, but not RATS 8.3

Use this forum to post questions about syntax problems or general programming issues. Questions on implementing a particular aspect of econometrics should go in "Econometrics Issues" below.
macro
Posts: 70
Joined: Thu Jul 16, 2015 3:01 pm

The STEPS instruction crashes in RATS 9, but not RATS 8.3

Unread post by macro »

This code

Code: Select all

calendar(q) 1969 1

compute begin_samp = 1969:1
compute end_samp   = 2006:4
compute max_lags = 4

open data "sandbox_data.xls"
    data(format = xls, org = columns)
close data
inquire(regressorlist) begin_est end_est
    # lgdpk{0 to max_lags} lcdk{0 to max_lags} ffedk{0 to max_lags}

system(model = mdl)
    variables lgdpk lcdk ffedk
    lags 1 to max_lags
    deterministic constant
end(system)
estimate(model = mdl) begin_est end_est

declare rectangular[series] residuals
steps(model = mdl, print, from = begin_est, to = end_est, residuals = residuals)
works in in RATS 8.3 Pro, but crashes RATS 9.0 Pro. Both of these versions are running on the same machine (Windows 7 x64). Was something changed in RATS between these versions that is causing this to fail? I looked in the reference manuals for each versions, and I can't find any syntax changes between versions that would cause the code to fail in 9.0.

Unfortunately, for this project, going back to RATS 8.3 isn't an option for me because I'm using some of the features that are new in RATS 9.0 (as discussed in other posts).
Attachments
sandbox_data.xls
(50.5 KiB) Downloaded 826 times
macro
Posts: 70
Joined: Thu Jul 16, 2015 3:01 pm

Re: The STEPS instruction crashes in RATS 9, but not RATS 8.

Unread post by macro »

I should add that the error in STEPS isn't a significant problem, because I believe I can use

Code: Select all

declare vector[series] residuals
forecast(model = mdl, print, static, from = begin_est, to = end_est, errors = residuals)
instead of

Code: Select all

declare rectangular[series] residuals
steps(model = mdl, print, from = begin_est, to = end_est, residuals = residuals)
to get the forecast errors (I realize that STEPS returns both the forecasts and the forecast errors in -residuals-, while FORECAST has separate options governing that output).
Post Reply