Generating correlations in Access

[Cblab]

Quick question - or probably not.

I'm doing a statistical exercise in which I need to generate mathematical
correlations between different rows of data. Specifically, I've got markets
in rows and combinations of 'driver' data in rows, and week #s in columns. I
want to determine which of the drivers correlations most strongly with the
market activity.

I can do this in Excel for a smaller number of rows, but I'm up to 40,000
rows now and Excel is barfing on this. Any way to do it in Access?

Thanks in advance!

(hope this is the right group)

 

[Tom Wickerath]

Hi Cblab,

If Excel is throwing fits with this much data, then I'd look into using
Total Access Statistics by FMS, if your budget can allow for the required
expenditure:

http://www.fmsinc.com/products/statistics/index.html

Of course, I'd try verifying with them first that their product would not
barf under similar conditions.

Tom

http://www.access.qbuilt.com/html/expert_contributors.html
http://www.access.qbuilt.com/html/search.html
__________________________________________

:

Quick question - or probably not.

I'm doing a statistical exercise in which I need to generate mathematical
correlations between different rows of data. Specifically, I've got markets
in rows and combinations of 'driver' data in rows, and week #s in columns. I
want to determine which of the drivers correlations most strongly with the
market activity.

I can do this in Excel for a smaller number of rows, but I'm up to 40,000
rows now and Excel is barfing on this. Any way to do it in Access?

Thanks in advance!

(hope this is the right group)

 

[Ron2005]

Access will probably not have any problem with the volume.

However it will probably require really knowing what you want and how
to figure it out from a different point of view. Without a clearer
idea of what the data looks like and the end result should look like I
(for one) can't say anything beyond what I have already.

Ron