noname presented the following explanation :
Well the SAS output is in the form of an HTML file.

in Internet Explorer it looks like this:

Number in
Model R-Square SSE Variables in Model
1 0.6484 189.36323 x1
1 0.5855 223.25558 x2
1 0.5047 266.78526 x3
1 0.4643 288.52191 x4
2 0.6716 176.88309 x1 x2
2 0.6671 179.33269 x1 x4
2 0.6628 181.60644 x1 x3
2 0.6165 206.56222 x2 x4
2 0.6014 214.71458 x2 x3
2 0.5394 248.07512 x3 x4
3 0.6803 172.18992 x1 x2 x4
3 0.6740 175.61857 x1 x2 x3
3 0.6701 177.68955 x1 x3 x4
3 0.6183 205.57403 x2 x3 x4
4 0.6854 174.17081 x1 x2 x3 x4

....
....
currently this file has 16 variables and their combinations, but the
variables could be more. hence the large size. i have to parse it in an array
like this and then calculate the Shapeley coeff & score for each variable
occurring with others & without others.

so a better design is required so that i can group and sum variable
combinations properly.

Now that I know this, you'll find what you need by googling "parse web
page in VB6/VBA". Best wishes...

--
Garry

Free usenet access at http://www.eternal-september.org
ClassicVB Users Regroup! comp.lang.basic.visual.misc