Last data update: 2014.03.03
|
R: Data Set for the Package Examples
Data Set for the Package Examples
Description
plmm.data is a data.frame of simulated, hierarchically structured (2-level) data used in examples provided in the package documents.
Usage
data(plmm.data)
Format
A data frame with 363 observations on the following 9 variables.
- y0
a numeric vector of the response simulated with one covariate for the nonparametric component under homoskedastic regression error assumption.
- y1
a numeric vector of the response simulated with one covariate for the nonparametric component under heteroskedastic regression error assumption.
- y2
a numeric vector the response simulated with two covariates for the nonparametric component under homoskedastic regression error assumption.
- x1
a factor with levels {F , M }.
- x2
a numeric vector of a covariate in the fixed parametric component.
- x3
a numeric vector of a covariate in the fixed parametric component.
- t1
a numeric vector of a covariate in the fixed nonparametric component.
- t2
a numeric vector of a covariate in the fixed nonparametric component.
- cluster
a clustering variable. There are 20 clusters. The cluster size is unbalanced; each cluster consists of from 5 to 30 observations.
Details
Data are generated from the following three models:
y0=β_0+β_1x1+β_2x2+β_3x3+γ_1(t1)+u+e_0
y1=β_0+β_1x1+β_2x2+β_3x3+γ_1(t1)+u+e_1
y2=β_0+β_1x1+β_2x2+β_3x3+γ_2(t1,t2)+u+e_1
where
(β_0, β_1, β_2, β_3)=(1, 1, 0.5, -1)
γ_1(t1)=2sin(t1)
γ_2(t1, t2)=2sin(t1)+0.3t2+0.75t2^2
u sim N(0,3)
e_0 sim N(0,1)
e_1=x3*e_0
u is the cluster-specific random intercept. x2 and t1 are correlated with correlation coefficient about 0.67.
Examples
data(plmm.data)
str(plmm.data)
Results
|