://www.wvbauer.com/
Original Message
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Daniel Malter Sent:
Tuesday, February 26, 2008 08:25 To: 'r-help'
Subject: [R] OLS standard errors
> Hi,
>
> the standard errors of the coefficients in two regressions that I
>
Please check your statistical methods lecture notes. var(e) has divisor
n-1, and that is not an unbiased estimator of the residual variance when
'e' are residuals. From summary.lm (and you are allowed to read the code)
rdf <- n - p
if (is.na(z$df.residual) || rdf != z$df.residual)
Daniel Malter wrote:
> Hi,
>
> the standard errors of the coefficients in two regressions that I computed
> by hand and using lm() differ by about 1%. Can somebody help me to identify
> the source of this difference? The coefficient estimates are the same, but
> the standard errors differ.
>
> ###
Hi,
the standard errors of the coefficients in two regressions that I computed
by hand and using lm() differ by about 1%. Can somebody help me to identify
the source of this difference? The coefficient estimates are the same, but
the standard errors differ.
Simulate data
hap
4 matches
Mail list logo
