[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Pearson's correlation coefficient

From   Austin Nichols <[email protected]>
To   [email protected]
Subject   Re: st: RE: Pearson's correlation coefficient
Date   Wed, 13 Jan 2010 14:51:45 -0500

Martin Weiss <[email protected]>:
Note that the sig option does not give a standard error on the
correlation.  Instead, it gives the p value from a test of b=0 in a
linear regression, which is also a test of zero correlation.

sysuse auto, clear
pwcorr headroom price, sig
estat vce
reg price headroom
estat vce
reg headroom price

For more interesting hypotheses, the problem is harder.
See also e.g.
for some relevant discussion.

On Wed, Jan 13, 2010 at 2:12 PM, Martin Weiss <[email protected]> wrote:
> <>
> *******
> sysuse auto, clear
> pwcorr mpg price, sig
> *******
> Martin
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Sibel S.
> Sent: Mittwoch, 13. Januar 2010 20:04
> To: [email protected]
> Subject: st: Pearson's correlation coefficient
> I am trying to measure standardized persistence between children's
> years of schooling and parents' average years of schooling using
> Pearson's correlation coefficient. I looked through the FAQ, but could
> not find information on this.  Is there a command for this? I also
> need to obtain the standard errors for the correlation coefficients.
> Thanks very much in advance.
> Sibel

*   For searches and help try:

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