Actually, I missed the part where you wanted to get a cumulative sum, rather just the sum.
I am pretty sure there is smarter and more efficient ways to do this. This will work though.
I am essentially creating the raw data as well, for you to test and play with
// define values
Col1Vals = {"a","a","a","a","b","b","c","c","c"};
dt = New Table();
dt << New Column("MyValues",Character,Nominal,<< Set Values(Col1Vals));
// Generate summary
dt1 = dt << Summary( Group( :MyValues ), Freq( "None" ), Weight( "None" ) );
UniqueItems = dt1:MyValues << Get values;
FrequencyOfItems = dt1 << Get As Matrix;
Close(dt1,"No Save");
// Add Cumulative Frequencies back
dt << New Column("CumulativeFreqCount",Numeric,Continuous);
for(i = 1, i <= N Items(UniqueItems),i++,
Rows = dt << Select Where(:MyValues == UniqueItems[i]) << Get Selected Rows;
ValsToAppend = Transpose(Index(1,FrequencyOfItems[i]));
dt[Rows,"CumulativeFreqCount"] = ValsToAppend;
dt << Clear Select;
);
Best
Uday