From: Daniel Farewell <farewelld_at_Cardiff.ac.uk>

Date: Tue 17 Jan 2006 - 22:17:21 EST

R-help@stat.math.ethz.ch mailing list

https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Tue Jan 17 23:03:56 2006

Date: Tue 17 Jan 2006 - 22:17:21 EST

Then a call to `glm' on the group 1 subset gives

> glm(y ~ x, family = poisson, data = df, subset = gp == 1)

Call: glm(formula = y ~ x, family = poisson, data = df, subset = gp == 1)

Coefficients:

(Intercept) x -1.0143 0.9726 Degrees of Freedom: 99 Total (i.e. Null); 98 Residual Null Deviance: 138.5 Residual Deviance: 82.76 AIC: 178.5

(the right answer) but `lmList' gives

> show(lmList(y ~ x | gp, family = poisson, data = df))

Call: lmList(formula = y ~ x | gp, data = df, family = poisson)
Coefficients:

(Intercept) x

1 0.5560377 0.6362124 2 1.8431794 1.8541193 3 4.5773106 4.7871929

Degrees of freedom: 300 total; 294 residual Residual standard error: 2.655714

which come from linear model fits, e.g.

> lm(y ~ x, data = df, subset = gp == 1)

Call:

lm(formula = y ~ x, data = df, subset = gp == 1)

Coefficients:

(Intercept) x 0.5560 0.6362

Any suggestions as to why lmList matches the linear fits rather than the GLM fits would be greatly appreciated. I'm using R2.2.1 with lme version 0.98-1 in Windows XP.

Daniel Farewell

Cardiff University

R-help@stat.math.ethz.ch mailing list

https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Tue Jan 17 23:03:56 2006

*
This archive was generated by hypermail 2.1.8
: Fri 03 Mar 2006 - 03:42:04 EST
*