An approximated principal component prediction model for continuous-time stochastic processes

Abstract

In this paper, a linear model for forecasting a continuous‐time stochastic process in a future interval in terms of its evolution in a past interval is developed. This model is based on linear regression of the principal components in the future against the principal components in the past. In order to approximate the principal factors from discrete observations of a set of regular sample paths, cubic spline interpolation is used. An application for forecasting tourism evolution in Granada is also included. © 1997 by John Wiley & Sons, Ltd.

Composition analysis of glass products based on correlation analysis

“…It can be seen that the correlation degree between lead oxide and barium oxide is the highest. By referring to the [3] and the known subject information, it can be seen that lead-barium glass contains more lead oxide and barium oxide, which is consistent with the conclusion of this section [10,11].…”

Functional Principal Component Regression and Functional Partial Least‐squares Regression: An Overview and a Comparative Study

“…This result suggests that selecting the scores with larger squared correlations with the scalar response may be more appropriate than selecting the scores linked with the largest eigenvalues, an idea hinted by Aguilera et al. () in the problem of predicting continuous‐time stochastic processes. Hence, we also consider an estimate of β that takes into account the response when choosing the scores to include in the regression.…”

Intraday Functional PCA Forecasting of Cryptocurrency Returns

“…A daily function of returns may as well start at a different point and cover 24 consecutive hours. This idea underlies the forecasting approach based on a "rolling FPCA" and is inspired by the paper of Aguilera et al (1997).…”