Can't get LinEst to work in VBA with an array.

Hello,
This is part of my code below, Correl works for the same array, but LinEst doesn't. I have noticed that LineEst keeps changing to lower case automatically. I am using Excel 2003. I have tried looking up other similar issues, but there isn't enough information in any of the posts I read, as to what the problem was.
Can anyone see why this is not working?
Thanks,

Please Login or Register  to view this content.

Please follow forum rules and use code tags when posting code

This forum can be very strict about putting your code within code tags. You should edit your post to include code tags, to avoid getting your topic locked.
I cannot look at this now (don't have Excel on this computer), but I will try to look at it later. In the meantime, can you provide any details beyond "it doesn't work"? Does it give you a compile error? runtime error? runs but the regresion is incorrect? Have you stepped through the code and identified the statement where it goes wrong?

Your arrays are not initialized in the code you posted.

I cannot recreate the error. As shg noted, the code block you posted does not tell us what is in p and w. If I assume something reasonable and simple for these values, then the code runs without error. For example, if I change the arrP and arrW assignmenet statements:

Please Login or Register  to view this content.

The code runs without error and correctly gives me slope=2 and intercept=0.

I don't think the problem is with the LinEst statement. It appears to me that the problem is something in the data being fed to the Linest statement. At this point, I would suggest that you use the locals window and or watch window and or the immediate window (see here http://www.cpearson.com/Excel/DebuggingVBA.aspx or other page if you are unfamiliar with these debugging tools) to see what is stored in p(), w(), arrp, and arrw to identify what is wrong with the input data.

Off the top of my head, things that Correl() will handle that LinEst() will not would be things like blanks, text, or other non-numeric data.