Activity Number:
|
546
- Recent Advances in Time Series and Point Process
|
Type:
|
Invited
|
Date/Time:
|
Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract #300556
|
|
Title:
|
Time Series Forecasting with Random Forests and Nonparametric Models
|
Author(s):
|
Barbara Ann Bailey*
|
Companies:
|
San Diego State University
|
Keywords:
|
ensemble method
|
Abstract:
|
The ensemble statistical learning method of random forests has successfully been used for prediction in wide range of applications. Random forests consist of an ensemble of decision trees for regression. Each tree in the ensemble is grown using a bootstrap sample of the data. In addition, when growing a tree each node is split using the best among a subset of predictors randomly chosen at that node. For the final prediction, the class mean prediction from all trees are used. We investigate the use of random forests for the modeling and forecasting of time series data. The stationary bootstrap is implemented to generate realizations of the time series to be used in the building of each tree in the random forest and in the construction of forecast intervals.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2019 program
|