createZforSplines {ecpc} | R Documentation |
Create a co-data matrix Z of splines
Description
Create a co-data matrix Z of spline basis functions for a continuous co-data variable.
Usage
createZforSplines(values, G=10, bdeg=3, index=NULL, p=NULL)
Arguments
values |
A vector with continuous co-data values. |
G |
Number of B-splines. |
bdeg |
Degree of the B-spline basis functions. |
index |
Index of the covariates corresponding to the values supplied. Useful when part of the co-data is missing/seperated and only the non-missing/remaining part should be modelled with splines. |
p |
Number of covariates in total. If not given, taken as length of ‘values’. But in cases where some covariates are left unpenalised, the total number of covariates may be larger. |
Value
A (pxG)-dimensional co-data matrix.
References
See for an introduction on p-splines:
Eilers, P. H., & Marx, B. D. (2021). Practical Smoothing: The Joys of P-splines. Cambridge University Press.
See Also
Use createS
to create a difference penalty for p-splines.
Examples
#create co-data with random normally distributed values for 100 covariates
values <- rnorm(n=100)
#suppose that there is one additional covariate (the first) that should not be modelled
ind <- 2:101
p<-101
Z <- createZforSplines(values=values,G=10,index=ind,p=p)