Actions for How to use vector autoregressive (VAR) models

Email RIS file
Bookmark
Report an Issue

How to use vector autoregressive (VAR) models / Shalini Aggarwal, Sunita Arora, Arvinder Kaur

Author
Aggarwal, Shalini
Published
London : SAGE Publications Ltd, 2025.
Physical Description
1 online resource : illustrations
Additional Creators
Arora, Sunita and Kaur, Arvinder
Access Online

Availability

I Want It
I Want It

Finding items...

Summary
When data are arranged chronologically, they are called time-series data. Time-series data have some unique features. For example, they may be correlated with their previous term, a feature called autocorrelation. Another feature is that data are nonstationary (i.e., their mean, variance, or covariance changes over time). Analyzing time-series data requires checking for stationarity and then deciding which of the analysis techniques should be applied to the data. One such technique is vector autoregression (VAR). VAR is applied when two or more time series are stationary at level or are stationary at first difference but are not cointegrated. VAR establishes lead-lag relationships and is useful to decide whether one variable is significant in forecasting the other variable. This How-to Guide describes when to apply VAR and how to interpret the results and illustrates the model's generalization. Furthermore, it describes the steps to build a VAR model-which include checking the stationarity of the variables, model selection, lag-length selection, and interpretation of the results-and shows how the VAR model is built in the EViews software package to analyze the relationship between the prices and return of four currencies: USD, yen, GBP, and euro.
Subject(s)
ISBN
9781036213596 : SAGE Research Methods: Business
View MARC record | catkey: 47380578