Predicting the present with Bayesian structural time series (Q90324): Difference between revisions
From MaRDI portal
Changed an Item |
Normalize DOI. |
||
(7 intermediate revisions by 7 users not shown) | |||
Property / DOI | |||
Property / DOI: 10.1504/ijmmno.2014.059942 / rank | |||
Property / DOI | |||
Property / DOI: 10.1504/IJMMNO.2014.059942 / rank | |||
Property / author | |||
Property / author: Steven L. Scott / rank | |||
Property / author | |||
Property / author: Steven L. Scott / rank | |||
Normal rank | |||
Property / review text | |||
Summary: This article describes a system for short term forecasting based on an ensemble prediction that averages over different combinations of predictors. The system combines a structural time series model for the target series with a regression component capturing the contributions of contemporaneous search query data. A spike-and-slab prior on the regression coefficients induces sparsity, dramatically reducing the size of the regression problem. Our system averages over potential contributions from a very large set of models and gives easily digested reports of which coefficients are likely to be important. We illustrate with applications to initial claims for unemployment benefits and to retail sales. Although our exposition focuses on using search engine data to forecast economic time series, the underlying statistical methods can be applied to more general short term forecasting with large numbers of contemporaneous predictors. | |||
Property / review text: Summary: This article describes a system for short term forecasting based on an ensemble prediction that averages over different combinations of predictors. The system combines a structural time series model for the target series with a regression component capturing the contributions of contemporaneous search query data. A spike-and-slab prior on the regression coefficients induces sparsity, dramatically reducing the size of the regression problem. Our system averages over potential contributions from a very large set of models and gives easily digested reports of which coefficients are likely to be important. We illustrate with applications to initial claims for unemployment benefits and to retail sales. Although our exposition focuses on using search engine data to forecast economic time series, the underlying statistical methods can be applied to more general short term forecasting with large numbers of contemporaneous predictors. / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 62P20 / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 62M20 / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 91B84 / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 65C40 / rank | |||
Normal rank | |||
Property / zbMATH DE Number | |||
Property / zbMATH DE Number: 6379195 / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
Bayesian model averaging | |||
Property / zbMATH Keywords: Bayesian model averaging / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
Bayesian structural time series models | |||
Property / zbMATH Keywords: Bayesian structural time series models / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
Markov chain Monte Carlo | |||
Property / zbMATH Keywords: Markov chain Monte Carlo / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
economic time series | |||
Property / zbMATH Keywords: economic time series / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
machine learning | |||
Property / zbMATH Keywords: machine learning / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
predicting the present | |||
Property / zbMATH Keywords: predicting the present / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
spike and slab priors | |||
Property / zbMATH Keywords: spike and slab priors / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
state space models | |||
Property / zbMATH Keywords: state space models / rank | |||
Normal rank | |||
Property / Wikidata QID | |||
Property / Wikidata QID: Q114106132 / rank | |||
Normal rank | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Property / DOI | |||
Property / DOI: 10.1504/IJMMNO.2014.059942 / rank | |||
Normal rank | |||
links / mardi / name | links / mardi / name | ||
Latest revision as of 11:16, 9 December 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Predicting the present with Bayesian structural time series |
scientific article |
Statements
5
0 references
1/2
0 references
4
0 references
2014
0 references
10 December 2014
0 references
Predicting the present with Bayesian structural time series (English)
0 references
Summary: This article describes a system for short term forecasting based on an ensemble prediction that averages over different combinations of predictors. The system combines a structural time series model for the target series with a regression component capturing the contributions of contemporaneous search query data. A spike-and-slab prior on the regression coefficients induces sparsity, dramatically reducing the size of the regression problem. Our system averages over potential contributions from a very large set of models and gives easily digested reports of which coefficients are likely to be important. We illustrate with applications to initial claims for unemployment benefits and to retail sales. Although our exposition focuses on using search engine data to forecast economic time series, the underlying statistical methods can be applied to more general short term forecasting with large numbers of contemporaneous predictors.
0 references
Bayesian model averaging
0 references
Bayesian structural time series models
0 references
Markov chain Monte Carlo
0 references
economic time series
0 references
machine learning
0 references
predicting the present
0 references
spike and slab priors
0 references
state space models
0 references