European Journal of Statistics and Probability (EJSP)

EA Journals

Forecasting Climatic Variables using Vector Autoregression (VAR) Model

Abstract

Although it is a topic of global concern, climate change implementation is typically regional. Finding an adequate Vector Autoregression (VAR) model to predict temperature, rainfall, and cloud coverage for the Jessore region of Bangladesh was the goal of this research project. The stationarity of variables was determined by ADF, PP, and KPSS unit root tests. Granger causality test was used to verify the endogenity among the variables. Employing AIC, VAR (11) model found best. The parameters associated with the model were estimated using the ordinary least square approach. Forecast error variance decomposition and impulse response function were utilized to reveal structural analysis, and the outcome revealed endogenous in the future. The predicted value showed a trend toward increasing temperature and a trend toward decreasing rainfall and cloud coverage.

Keywords: Granger Causality, forecast error variance (FEV) decomposition, impulse response function, vector autoregression (VAR), white noise

cc logo

This work by European American Journals is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 Unported License

 

Recent Publications

Email ID: submission@ea-journals.org
Impact Factor: 6.90
Print ISSN: 2055-0154
Online ISSN: 2055-0162
DOI: https://doi.org/10.37745/ejsp.2013

Author Guidelines
Submit Papers
Review Status

 

Scroll to Top

Don't miss any Call For Paper update from EA Journals

Fill up the form below and get notified everytime we call for new submissions for our journals.