A Comprehensive Review and Analysis for forecasting Industrial Data

The manuscript represents multiple methods of forecasting real multivariate time series dataset. The dataset related to Annual Survey of Industries is collected for the analysis pupose. What makes the forecasting of multivariate data complicated are complex and non-linear interdependencies between the components of time series data. So it is crucial to make predictions accurately and model long term dependency in data. The vector autoregression (VAR) and vector error correction (VEC) models are two models that have proven to be most common and useful models in illustrating dynamic behavior of the time series analysis and its long term forecasting.