Forecasting time series by functional PCA. Discussion of several weighted approaches (Q1979102)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Forecasting time series by functional PCA. Discussion of several weighted approaches |
scientific article |
Statements
Forecasting time series by functional PCA. Discussion of several weighted approaches (English)
0 references
24 May 2000
0 references
The authors study continuous time series that have been observed only at discrete time points. A weighted estimation of the principal components is applied to forecast such time series. Several approaches are used to model and forecast a simulated narrow-band process by using different types of weighting functions. The model is compared with multivariate principal component regression models.
0 references
time series
0 references
principal components
0 references
orthogonal expansions
0 references
weighted functional estimation
0 references
interpolating splines
0 references
