1986Working paperOpen access

Comparison of Univariate ARIMA, Multivariate ARIMA and Vector Autoregression Forecasting

Michael Bagshaw

Open full text 4 citations

Abstract

A comparison of the forecasting abilities of univariate ARIMA, multivariate ARIMA, and VAR, and examination of whether series should be differenced before estimating models for forecasting purposes.

Open-access reader

About this research paper

What this paper is about

A comparison of the forecasting abilities of univariate ARIMA, multivariate ARIMA, and VAR, and examination of whether series should be differenced before estimating models for forecasting purposes.

Why it matters

OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

A comparison of the forecasting abilities of univariate ARIMA, multivariate ARIMA, and VAR, and examination of whether series should be differenced before estimating models for forecasting purposes.

Key concepts: Autoregressive integrated moving average, Univariate, Multivariate statistics, Vector autoregression, Econometrics, Autoregressive model, Statistics, Time series

Related papers

Back to paper searchBrowse research topicsOriginal source
Comparison of Univariate ARIMA, Multivariate ARIMA and Vector Autoregression Forecasting — Research Paper | ScholarLens