Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Pearson correlation with check

From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: RE: Pearson correlation with check
Date   Thu, 23 Nov 2006 23:59:07 -0000

You can get a matrix of correlations using -correlate- 
or with -pwcorr-. Given many missing data, the results 
might be very different. 

You can get a printout of significance levels using -pwcorr-. 

Why you failed to find these commands, which are only a 
-search- away from you, is not clear. 

Getting significance results that mean anything much is quite 
another matter. As panel data, these data will not be mutually 
independent, so the standard P-values associated with Pearson
correlations do not apply. (The same issue would arise with any
other kind of correlation.) I doubt that you can calculate 
meaningful P-values without some explicit model for the data 
generating process. That's probably irrelevant given the large 
notional "sample size". 

[email protected] 

[email protected]
> I need an help: I have a panel dataset of 1000 firms over ten 
> years (1980-1990), a lot of variables (but the panel is 
> incomplete, because some data are missing), and I must 
> realize a table of Pearson correlations among measures of 
> var1, var2, var3. 
> The following should be the plot:
>          var1     var2     var3
> -----------------------------------
> var1
> var2
> var3
> Further, for each value I need to check if every correlation 
> is significantly different from zero at the 0.05 level or 
> higher. I don't know how to do the table and to check it... 
> Any help will be greatly appreciated.

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index