Automatically increase number of random cases selected with SPSS syntax and macros - spss

Question

I'm trying to force SPSS to do a psuedo-Monte Carlo study. The real world data is so bizarre that I can't reliably simulate it (if you're interested, it is for testing Injury Severity Scores). As such, I'm using a dataset of about 0.5 million observations of the real world data and then basically bootstrapping the results from increasingly large random samples of it. The goal is to figure out what group sizes are necessary to assume normality (at what group sizes do t-tests and Mann-Whitney U tests reliably agree; in other words, when can I count on the Central Limit Theorem).
My plan is to use a combination of a macro to repeat the two tests 100 times (but run 150 times in case the random selection results in a group size of zero), and then use OMS commands to export the results of the numerous tests into a separate data file.
So far, everything works just fine but, I would like to add another looping command to run the process again but select more random cases. So, it would run 150 times with 10 random cases selected each time, then, after running the first 150, it would run another 150 but select 20 random cases. Optimally, it would be something like this:
Select 10 random cases
Run a t-test and a Mann-Whitney U test
Repeat 150 times
Select 20 random cases
Run a t-test and a Mann-Whitney U test
Repeat 150 times
...
(After running on 200 cases, now increase by 50)
Select 250 random cases
Run a t-test and a Mann-Whitney U test
Repeat 150 times
Select 300 random cases
...
Select 800 random cases
Run a t-test and a Mann-Whitney U test
Repeat 150 times
(Stop after running on 800 cases)
Save all of these results using OMS
Everything in the below syntax works perfectly, except for one small issue, I can't figure out how to have it increase the size of the random sample, and I would prefer not to do that manually.
Even if I have to do it manually, is there a way to append the latest results to the existing file instead of replacing the existing file?
DEFINE !repeater().
!DO !i=1 !TO 150.
*repeat the below processes 150 times
*select a random sample from the dataset
DATASET ACTIVATE DataSet1.
USE ALL.
do if $casenum=1.
compute #s_$_1=10.
compute #s_$_2=565518.
* 565518 is the total number of cases
end if.
do if #s_$_2 > 0.
compute filter_$=uniform(1)* #s_$_2 < #s_$_1.
compute #s_$_1=#s_$_1 - filter_$.
compute #s_$_2=#s_$_2 - 1.
else.
compute filter_$=0.
end if.
VARIABLE LABELS filter_$ 'x random cases (SAMPLE)'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.
*run a non-parametric test
NPAR TESTS
/M-W= issloc BY TwoGroups(0 1)
/MISSING ANALYSIS.
*run a parametric test
T-TEST GROUPS=TwoGroups(0 1)
/MISSING=ANALYSIS
/VARIABLES=issloc
/CRITERIA=CI(.95).
!DOEND.
!ENDDEFINE.
*use OMS to extract the reported descriptives and results from the viewer
*and save them to a file
OMS /SELECT TABLES
/DESTINATION FORMAT = SAV OUTFILE = 'folder/folder/OMS file.sav'
/IF SUBTYPES=['Mann Whitney Ranks' 'Mann Whitney Test Statistics' 'Group Statistics' 'Independent Samples Test']
/COLUMNS SEQUENCE = [RALL CALL LALL].
!repeater.
OMSEND.

Answer 1

Never mind. The answer was so obvious, I missed it entirely. I just needed to define the sample size selection within the macro. *facepalm