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

Rice Production in Bangladesh Employing by Arima Model

M. A. AWAL
Senior Scientific Officer (Agricultural Economics)
On-Farm Research Division, Bangladesh Agricultural Research Institute (BARI), Mymensingh

M.A.B. SIDDIQUE
Senior Scientific Officer
Agricultural Economics Division, Bangladesh Rice Research Institute (BRRI), Gazipur 1701, Bangladesh

Abstract/ Executive Summary:

The present study was carried out to estimate growth pattern and also examine the best ARIMA model to efficiently forecasting Aus, Aman and Boro rice production in Bangladesh. It appeared that the time series data for Aus and Aman were 1st order homogenous stationary but Boro was 2nd order stationary. The study revealed that the best models were ARIMA (4,1,4), ARIMA (2,1,1), and ARIMA (2,2,3) for Aus, Aman, and Boro rice production, respectively. The analysis indicated that short-term forecasts were more efficient for ARIMA models compared to the deterministic models. The production uncertainty of rice could be minimized if production were forecasted well and necessary steps were taken against losses. The findings of this study would be more useful for policy makers, researchers as well as producers in order to forecast future national rice production more accurately in the short run.

Keywords: Production, ARIMA model and forecasting of rice

Location: BARI, Gazipur

Start Date:

End Date:

Program Area: Socio-economic and Policy

Commodity: Rice

Objectives:

i) To estimate the growth pattern of rice production of different seasons in Bangladesh.

ii) To develop appropriate ARIMA models for the time series of Aus, Aman, and Boro rice production in Bangladesh.

iii) To make five-year forecasts for all the time series with appropriate prediction interval; and compare the forecasting performance of ARIMA and deterministic models.

Methodology:

The present study was conducted mainly based on the secondary data. Majority of the information on the time series data of Aus, Aman, and Boro rice production pertaining to the period of 19972-73 to 2007-08 were generated and compiled from published volumes of Bangladesh Bureau of Statistics (BBS). lnferential statistical tools were employed in analyzing the data. In order to forecast the rice production was modeled by Box-Jenkins type stochastic autoregressive integrated moving average (ARIMA) process. The Box-Jenkins type ARIMA process (Box and Jenkins, 1978) can be defined as ø(B) (?d yt-µ) = θ(B)εt. Here, yt denotes Aus, Aman, and Boro rice production in million metric tons, µ is the mean of ?d Yt, ø(B) is 1-ø1B----øpBp , θ(B) is 1-θ1B----θqBq , θ denotes the moving average parameter, ø denotes the autoregressive parameter, p, q, and d denote the autoregressive, moving average and difference orders of the process, respectively, ? and B denote the difference and back-shift operators, respectively. The estimation methodology of the above model consists of three steps, namely identification, estimation of parameters, and diagnostic checking. The identification step involves the use of the techniques for determining the value of p, d, and q. Here, these values are determined by using autocorrelation and partial autocorrelation functions (ACE and PACF) and Augmented DickeyFuller (ADF) test. The model used for ADF is y-yt-1=α+βt+(ρ-1)yt-1+λ?yt-1 (Pindyck and Rubinfeld, 1991, p.461).The second step is to estimate the parameters of the model. Here, the method of maximum likelihood is used for this purpose. The third step is to check whether the chosen model fits the data reasonably well. For this reason, the residuals are examined to find out if they are white noise. To test if residuals are white noise, the ACE of residuals and the Ljung and Box (1978) statistic are used. In case of two or more competing models passing the diagnostic checks, the best model is selected using the criteria multiple R² , Root Mean Squared Error (RMSE), Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Mean Absolute Error (MAE), and Mean Absolute Percent Error (MAPE). Nine deterministic types’ growth models are also considered in this study for comprising the forecasting efficiency of stochastic models. The above mentioned model selection criteria are used to select the best deterministic model forecasting purpose.

Source of Information: Bangladesh J. Agril. Res. 36(1) : 51-62, March 2011, ISSN 0258-7122

Article Link (Online Journal):

Funding Source:

Total Budget (Tk.) :

Achievement/Findings:

This study made the best effort to develop a short run forecasting model of rice production in Bangladesh. The empirical analysis indicated that ARIMA (4,1,1), ARIMA(2,1,1), and ARIMA(2,2,3) were best fitted model for short run forecasting of Aus, Aman, and Boro rice production, respectively. However, the ARIMA model estimated that the rice production of Bangladesh were 1735, 11056, and 13538 thousand metric tons in 2008-09 and if the trend in present rice production to be continued, the total rice (Aus, Aman, and Boro) production of Bangladesh would be 1591, 11612, and 15541 thousand metric tons in the year 2012-2013. To increase rice production of Bangladesh, the government should put emphasis on adopting of high yielding rice varieties in three rice seasons and take proper input subsidiary programme in order to achieve food security. The production uncertainty of rice could be minimized if production were forecasted well and necessary steps were taken against losses. For this purposes, the findings of this study would be more useful for policy makers, researchers, and producers for future national rice production policies.

Knowledge Management: Journal