A fixed effect curvilinear model
The authors introduce an extension of a fixed effects model for principal components analysis (PCA) which builds smooth optimal transformations of the data and then derive graphical displays. This leads to a new computational approach replacing B-splines or I-splines by a spline smoother. The estimation is performed by a smoothing alternative least squares algorithm (SALSA) implemented in the software S+. Estimates of the risk functions are computed with the help of bootstrap resampling. The efficiency of the approach is tested on simulated data by comparing the true risk function values and their estimates. Results obtained by SAS and SALSA are compared.
- Functional principal components analysis via penalized rank one approximation
- Generalized principal component analysis with respect to instrumental variables via univariate spline transformations
- Principal components analysis of sampled functions
- Penalized Spline Models for Functional Principal Component Analysis
- Parametric Functional Principal Component Analysis
This page was built for publication: A fixed effect curvilinear model
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1965997)