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

Forecasting Area, Production and Yield of Onion in Bangladesh by Using Arima Model

Farhana Arefeen Mila
Corresponding author:
Department of Agribusiness, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Gazipur-1706, Bangladesh.

Mst. Tania Parvin
Department of Agribusiness, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Gazipur-1706, Bangladesh.

Abstract/ Executive Summary:

In Bangladesh, onion is a widely used spices both for preparing food and curing diseases as it has medicinal values. As the demand for onion is increasing day by day, it is necessary to make actual projections of onion for undertaking some policies based on it. Therefore, the study investigates the future changes in the area, yield and production of onion in Bangladesh by using the most popular Box-Jenkins methodology. The autoregressive integrated moving average model has been used to understand the pattern of change over a period of 57 years (1961 to 2017) as well as to forecast the changes in the upcoming years. Some information criteria (such as AIC, AICc and BIC) was considered for selecting the best-fitted models of each variable. The forecasted results showed an upward trend for all the variables considered in this study. It implies that the area of onion will increase from 193932.6 hectares in 2018 to 265770.9 hectares in 2027. Again, the amount of onion production will increase from 2073.61 M tons to 3574.06 M tons and for onion yield, it will rise from 10343.17 Kg/ha to 12988.02 kg/ha from 2018 to 2027. These predictions may help the government balancing the demand with the supply and also regulating the price of onion in the domestic markets of Bangladesh.

Keywords: Onion, Non stationarity, Forecasting, Auto regressive integrated moving average, Bangladesh

Location: In Bangladesh

Start Date:

End Date:

Program Area: Crop-Soil-Water Management

Commodity: Onion, Modeling

Objectives:

To investigates the future changes in the area, yield and production of onion in Bangladesh by using the most popular Box-Jenkins methodology.

Methodology:

The study used annual time series data of onion area, production and yield in Bangladesh over the period of 1961 to 2017. The data were collected from the website of Food and Agricultural Organization and Bangladesh Bureau of Statistics, Government of the Peoples Republic of Bangladesh. The data were analyzed using R-3.5.3 software. The Box Jenkins autoregressive integrated moving average (ARIMA) models were applied for the observed data. Finally, a forecast was made for the period of 2018 to 2027. In time series analysis, an autoregressive integrated moving average (ARIMA) is a generalization of an autoregressive moving average model. As there is no seasonal variation in the data, so the study considered nonseasonal ARIMA (p, d, q) models. The model has three parts. The Box Jenkins methodology is applied both for Univariate or Multivariate ARIMA models in order to find out the best-fitted model of the time series data [16]. There are four steps of the Box-Jenkins method. At first, it is required to check the variables if it is stationary or not. Unit Root test is applied for checking the stationarity. The augmented Dickey-Fuller (ADF) test and Kwiatkowski, Phillips, Schmidt and Shin (KPSS) test can be used to check the unit root and stationarity of the data. If the data is not stationary, then the data is transformed into stationary by differencing the original data series. The second step is to identify a tentative model to specify the appropriate values of p, d and q. The plots of the Auto Correlation Function (ACF) and Partial Auto Correlation Functions (PACF) help us to recognize the order of MA and AR process respectively. The third step is to estimate the parameters of the model by using the likelihood methods such as AIC, AICc and BIC, Finally, it is necessary to check the best-fitted model by testing the parameters and residuals of the chosen model. The residuals are tested through the ACF and PACF plots and the Ljung Box [17] statistics. The forecasted data can be checked the validity of the model through mean absolute percentage error (MAPE). After checking the diagnostics step, the forecast can be generated.

Source of Information: Asian Journal of Agricultural Extension, Economics & Sociology, 37(4): 1-12, 2019; Article no.AJAEES.53040; ISSN: 2320-7027

Article Link (Online Journal): DOI: 10.9734/AJAEES/2019/v37i430274

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

A forecasting of onion area, production and yield model were formulated. Based on the forecasting and validation results, the study showed that there will be lesser supply of onion in future compared to domestic demand of onion in Bangladesh. Because, at present the domestic demand for onion is 24 lakh tons that might be doubled after 10 years because of larger population growth. Thus, it needs to increase the yield for meeting the huge demand of onion in the country. To overcome this situation, the onion producers can change its cropping pattern by bringing additional area under onion production. The onion producers can adopt advance methods and technology for accelerating the yield of onion. Government can give subsidy and provide high yielding seeds of onion for the growers. Also government can provide better storage facility for reducing the post-harvest loss and reining the over price hike of onion through syndication of dishonest traders in Bangladesh. Again, the onion production can be increased through efficient use of existing resources. For that, adequate capital needs to invest in research and development related to spices crops in Bangladesh. In this way, the scenario analyses under this study may boost up the productivity and fulfill the domestic demand of onion in near future.

Knowledge Management: Journal