Journal of Food, Agriculture and Environment




Vol 18, Issue 2,2020
Online ISSN: 1459-0263


 The challenge of selecting the best forecasting model for a time series data


Author(s):

 Bhola Nath, D. S. Dhakre, K. A. Sarkar, D. Bhattacharya

Recieved Date: 2020-01-30, Accepted Date: 2020-03-25

Abstract:

Fitting of an appropriate model to an observed time series data for the purpose of efficient prediction is always a challenging task to the researchers. The practitioners of statistics in their first attempt always try to fit a parametric regression model to the data observed. For all parametric models to be fitted, it is assumed that the model errors should follow independent normal distribution strictly or their distribution should be known. If that assumption on error distribution is not satisfied sometimes, then we should search for an alternative procedure of modelling such type of data. Here, we propose the nonparametric regression procedure as the alternative approach and try to study its performance. In this investigation the secondary data on production of rice  for the Kharif season and production of wheat  for Rabi season for India as a whole for 51 years (1962-1963 to 2012-2013) have been used. It has been found that the variable, production of rice, does not hold the assumption of normal distribution of errors but the variable, production of wheat satisfies this assumption of normality of error distribution. Here we have applied parametric and nonparametric regression and spline regression approaches to both the data sets. It has been observed that there is a great decrease in the value of Mean Absolute Percentage Error (MAPE) of the prediction for the dependent variable, i.e., production of rice when nonparametric regression is used. It is concluded that the nonparametric regression works pretty well for the data set for which the normality assumption of the error distribution does not satisfy and gives better prediction than traditional parametric regression. If data set contain a large number of observations, then spline regression fits the data well.

Keywords:

Assumptions, exponential fitting, MAPE, nonparametric regression, normal distribution, parametric regression, spline regression


Journal: Journal of Food, Agriculture and Environment
Year: 2020
Volume: 18
Issue: 2
Category: Environment
Pages: 97-102


Full text for Subscribers
Information:

Note to users

The requested document is freely available only to subscribers / registered users with an online subscription to this Journal. If you have set up a personal subscription to this title please enter your user name and password (https://www.wflpublisher.com/Pages/subscription-procedure). All abstracts are available for free.

Article purchasing

If you like to purchase this specific document such as article, visit the site https://www.wflpublisher.com/Journal. Software and compilation, Science & Technology, all rights reserved. Your use of this website details or service is governed by terms of use. Authors are invited to check from time to time news or information.


Purchase this Article:   20 Purchase PDF Order Reprints for 15

Share this article :