Improving summary() function

3 replies [Last post]
mspiegel's picture
Offline
Joined: 07/31/2009

The summary() function has been reimplemented. It now ignores columns in a dataset that are not used in a FIML objective function. It also now handles independent submodels more intelligently. The old implementation ran in about 10 seconds for models/failing/OneFactorCovRAMSpeedup.R (with numberIndicators = 80). The new implementation runs in about 0.5 seconds.

We have only three tests in the test suite that check the summary() output. Please get into the habit of adding omxCheckEquals() statements for the summary() output in addition to the parameter estimates and the other things we usually check in the test suite. It will help in the long term as we handle corner cases with the summary output.

Ryne's picture
Offline
Joined: 07/31/2009
Thanks, Michael! Is there

Thanks, Michael! Is there something specific you want us to write into omxCheckEquals tests of the summary function, or should we just pick a parameter or one component of a list to check?

mspiegel's picture
Offline
Joined: 07/31/2009
Oops. Yes, it's the

Oops. Yes, it's the observedStatistics and estimatedParameters components of the summary() output that concerns me. If we get those two correct, then we are likely to get the remaining components correct as well.

tbates's picture
Offline
Joined: 07/31/2009
nice speed up michael !

nice speed up michael !