Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Modeling and Forecasting Rice Production in Bangladesh: an Econometric Analysis

M.M. Miah*
Department of Statistics, Noakhali Science and Technology University, Sonapur, University Road, Bangladesh

Abstract/ Executive Summary:

The aim of this research is to find a time series model and to forecast the rice production of Bangladesh. That is achieved by finding the tentative Autoregressive Integrated Moving Average (ARIMA) models that fit and forecast well for rice production in Bangladesh. There are mainly three types of rice that are cultivated in Bangladesh throughout the year as Aus, Aman and Boro. Season covering the whole country we have included all these three types of rice production for analysis. In this paper, we study the performance of ARIMA model. We apply the time series Autoregressive Integrated Moving Average (ARIMA) model of different lag order to model rice production in Bangladesh. The suitable and efficient model to represent the data of the time series is chosen according to the smallest values of AIC, BIC, RMSE and MAPE criteria are used to select the best ARIMA (p,d,q) model. The result shows that, the best selected ARIMA model for Aus, Aman and Boro productions is ARIMA (2, 1, 5), ARIMA (2, 1, 5) and ARIMA (1, 1, 1) respectively than other tentative ARIMA (p,d,q) models. We forecasted the next seven years' rice production and make a comparison between the original series and forecasted series which also shows the same manner indicating fitted model are statistically well behaved to forecast rice productions in Bangladesh. It is found from the analysis that ARIMA model gives better forecasts more accurately in the short run.

Keywords: ARIMA, Rice Production, Bangladesh

Location:

Start Date:

End Date:

Program Area: Socio-economic and Policy

Non-commodity: Modeling

Objectives:

Modeling and forecasting the rice production is important because it helps the planners and the government to take the proper budget.

Methodology:

Data: Secondary data were collected from Bangladesh Agricultural Ministry’s website. The yearly production (000’ton) data are collected from this website for the period 1972 to 2015. ARIMA Model Autoregressive Integrated Moving Average (ARIMA) models intend to describe the current behavior of variables in terms of linear relationships with their past values. These models are also called Box-Jenkins (1984) models on the basis of these authors’ pioneering work regarding time-series forecasting techniques. An ARIMA model can be decomposed into two parts. First, it has an Integrated (I) component (d), which represents the amount of differencing to be performed on the series to make it stationary. The second component of an ARIMA consists of an ARMA model for the series rendered stationary through differentiation. The ARMA component is further decomposed into AR and MA components. The autoregressive (AR) component captures the correlation between the current value of the time series and some of its past values. Box and Jenkins (1970)[18]introduced the ARIMA model. It also referred to as Box-Jenkins methodology composed of a set of activities for identifying, estimating and diagnosing ARIMA models with time-series data. The model is most prominent method in financial forecasting. ARIMA models have shown the efficient capability to generate short-term forecasts. It constantly outperformed complex structural models in short-term prediction. An ARIMA (p, d, q) model, the future value of a variable is a linear combination of past values and past errors. Estimation of parameters: the second step is the estimation of the model parameters for the tentative models that have been selected. Diagnostic Tests of Residuals: The estimated model must be checked to verify if it adequately represents the series. The best model was selected based on the minimum values of Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Normalized Bayesian Information Criterion (BIC), Akaike Information Criterion (AIC). Diagnostic checks are performed on the residuals to see if they are randomly and normally distributed. Jarque-Bera test We can check the normality assumption using Jarque-Bera (1978) test, which is a goodness of fit measure of departure from normality, based on the sample kurtosis(k) and skewness(s).

Source of Information: Research & Reviews: Journal of Statistics ISSN: 2278-2273 (Online), ISSN: 2348-7909 (Print) Volume 8, Issue 2

Article Link (Online Journal):

Funding Source:

1.

Budget:

Total Budget (Tk.) :

Achievement/Findings:

We employed the Box-Jenkins approach to model and forecast rice production in Bangladesh. Our forecast showed an increasing pattern for Aman and Boro rice production and it is a little bit of fluctuation (i.e., increasing and decreasing) pattern for Aus rice production. In this research, we tried to fit the best model to forecast different types of seasonal rice productions in Bangladesh named as Aus, Aman and Boro. Model selection criteria such as AIC, BIC, RMSE and MAPE are used to select the best model. The best selected ARIMA model for forecasting Aus production is ARIMA (2, 1, 5), for Aman it is ARIMA (2, 1, 5), and for Boro it is ARIMA (1, 1, 1). Again, for each selected model the histogram of residuals showed that the residuals of these three models seem to be normally distributed with mean zero and constant variance. Hence these three models provide an adequate predictive model for the Aus, Aman and Boro rice production in Bangladesh. These models can help the researchers, policymakers, and Businessmen for future planning.

Knowledge Management: Journal