I'm hoping that this is a relatively easy question for someone familiar with
the lme4 package.
I'm accustomed to using HLM software and writing a simple 2 level [null]
equation like this:
L1 - Yij = b0 + e
L2 - b0 = B00 + u0
The following command in R provides results that are identical to the HLM
program.
results <- lmer( Y ~ 1 |id , PanelData4)
I can't seem to find any examples on-line nor in the help about how to write
the lmer4 formula that contains two predictor variables at level 1 with
fixed slopes.
L1 - Yij = b0 + b1(x) + b2(z) + e
L2 - b0 = B00 + u0
b1 = B10
b2 = B20
Can someone give me an example?
Thanks
Jeff
lme4 / HLM question
2 messages · Jeff, Ben Bolker
Jeff <r <at> jp.pair.com> writes:
I'm hoping that this is a relatively easy question for someone familiar with the lme4 package. I'm accustomed to using HLM software and writing a simple 2 level [null] equation like this:
L1 - Yij = b0 + e L2 - b0 = B00 + u0
The following command in R provides results that are identical to the HLM program.
results <- lmer( Y ~ 1 |id , PanelData4)
I can't seem to find any examples on-line nor in the help about how to write the lmer4 formula that contains two predictor variables at level 1 with fixed slopes.
L1 - Yij = b0 + b1(x) + b2(z) + e
L2 - b0 = B00 + u0
b1 = B10
b2 = B20
In general you'd probably be better off asking this question
at r-sig-mixed-models at r-project.org ... but it's very easy to
put fixed effects into a model --
lmer (Y ~ x + z + (1|id), PanelData4)
or in lme (nlme package):
lme ( Y~ x + z, random = ~ 1|id, PanelData4)
Neither lmer nor lme need to be told explicitly which level
the fixed effects vary at (the data are always provided in
long form).
Ben Bolker