[R-sig-ME] (no subject)

[Ben Bolker]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 15-03-09 06:53 AM, mervat mohamed via R-sig-mixed-models wrote:
> Dear lmer() friends,
> 
> I amtrying to estimate a model about regress deviance y on deviance
> x
> 
> Doesanyone of you has already done this before using lmer()
> method? i want the code of this model
> 
> Thanksa lot
> 
> mervat
> 
> 
> 

transcribing your equation into TeX format:

(y_{ij} - \bar y_{.j}) =
   \beta(x_{ij}-\bar x_{.j}) + e_{ij} - \bar e_{.j}

It's not really clear to me what you want to do.  It looks like
it *might* be a random_slope model.  In other words,

  y ~ x + (x|f)

where x is a continuous covariate and f is a grouping variable
(factor) gives

 y_{ij} = beta_0 + beta_1*x + (b_{0j} + b_{1j}*x)

Where b_{0j} is a group-level deviate from the intercept and b_{1j} is
a group-level deviation from the slope.

Is that translatable into your context?

Depending on how you want your results presented, van de Pol and
Wright 2009 might be useful:
http://www.myscience.eu/martijn/pdfs/vandePol%26Wright2009_AnimBeh.pdf

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iQEcBAEBAgAGBQJU/gtxAAoJEOCV5YRblxUHWpkH+wXX9P8xmK8pQoBmix/Jtk4E
MmS5V2G0REuuiZHcOpjNVfWH4+Mda3VrULiZ5C2PJkLAizAZ0ME/HTtW8Ai2wAGx
tJDlo5A19/WMGGo3OHH8hU2UroCJWFfa3Ion4jVMP6911t5lL4Eewb+B+BcxcvQD
QIhs8hH+i9Pv9lOEWtzpgxp7Elhjs0q4wXRF++CopBG3mKHtutl7G1+zFS+jm64o
ZztJ7MZ/oeAvNHTExi0rxdB6MDrcX5eXHrRiDCP5OD7rD6MWxfEezne1QV/V1feX
R21iN64wBiSSZmTFCy3xP+Vl7CZ/hJvXsKvPf7c0HwfUwN0TflXN3GO8Jf2Ebcs=
=iics
-----END PGP SIGNATURE-----