[R] (no subject)

[Thomas W Blackwell]

Silika  -

By far the best reference for the nlme package is the book by its
authors:

J.C. Pinheiro and D.M. Bates.  Mixed effects models in S and Splus.
Springer, NY, 2000.  ISBN (US) 0-387-989579.  (There is a different,
European, ISBN number but my library catalog doesn't give it.)

This citation is NOT given in  help("lme"),  presumably because the
help page pre-dates the book.  The Laird and Ware (1982) article
cited on the help page may also be useful.

-  tom blackwell  -  u michigan medical school  -  ann arbor  -

On Sat, 4 Oct 2003 silika.tereshchenko at unisg.ch wrote:

> Dear all,
>
> I have the following question. I have to fit the hierarchical model for the
> hypothesis concern the individual-level effects by controlling for the
> individual -level attributes and national-level contextual effects  on
> individuals by using R.
>
> O have to obtain the estimates of the impact of the second-level (national:
> GDP per capita) effects on individuals ( in this instance the impact of the
> GDP per capita on the attitudes towards the EU enlargement) by allowing
> national differences in both slopes (GDP per capita) and interceps.
>
> In R programm for the fitting the hierarchical models i can use the nlme
> package. I found a literature (Bryk and Raudenbush) for the hierarchical
> models and understood how to build this models by using the survey data.
> The question arise if I´m thinking about the combination of the datasets:
> the GDP per capita that will be hold as a constant and the survey data for
> each respondent.
>
> My question is how I could solve this problem and merge the data in such a
> form that should be appropriate for the hieararchical regression model.
>
> Thank a lot in advance for the help,
>
> Silika Tereshchenko