'codebook' throws error with complicated missings and large samples

[Garret Christensen]

I'm attempting to summarize and review a new dataset using 'codebook' and am encountering an error. Specifically, certain variables throw an "options not allowed" r(101) error, despite my not (knowingly) using any options. Example code produces the same error (using my Stata 14 SE on Linux).

clear all
set obs 100
gen x=4
codebook
replace x=.a in 1
codebook
replace x=.b in 2
codebook 
replace x=.m in 3
codebook
replace x=.z in 4
codebook
expand 300 
codebook

It seems to be some combination of data size and the multiple different missing values; I can't seem to reproduce the error with one of those alone. Any idea what's causing this? Perhaps some memory or matsize error? But the demo dataset isn't very large...

[Clyde Schechter]

I cannot replicate your problem on my setup, which uses Stata version 15.1 on Windows.

While that doesn't solve your problem, it at least verifies that there isn't something wrong with your code. In a situation like this I would start with the following:

1. Reboot your computer and try it again.
2. If that doesn't work, update your Stata. Run -update all, force-. An error like this one might arise from your ado files and executable being out of sync or something like that.
3. If that doesn't work, try uninstalling Stata and doing a fresh install.
4. If that doesn't work, contact Tech Support.

[Garret Christensen]

My personal 15.1 Windows installation doesn't throw the error either. If only it were that easy to get IT to reinstall software on my large organization's server

[Clyde Schechter]

Ugh! That's really unfortunate. I've been in that position. It's why I purchase my own Stata license and run it on my own computer: it's too "mission critical" to my work to let IT be in control of it!