Predictive Modeling on Time Series Data With Large Number of Predictors
Time series modeling is widely used in many areas, such as finance, economics, marketing and social science, in modern society. Because of the development of information technology, more and more data are available to users with prices lower than ever. This encourages forecast services users to consider predictive models for their decision-making processes and on the flip side, it brings challenges to traditional time series modeling procedures which can only handle limited number of predictors/covariates in general. The modeling methods in the field of machine learning can be applied to time series data but it is not as intuitive. This project will design a time-series model based method for data with a large number of predictors/covariates in a forecast combination frame.
Return to this PI's main page.