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

Exploring the Climate Change Impact on Major Food Crops of Bangladesh - a Time Series Analysis

Md. Mursalin Hossain Rabbi
MSS in Economics, Bangladesh University of Professionals, Bangladesh.

Nazia Tabassum
MSS in Economics, Bangladesh University of Professionals, Bangladesh.

Abstract/ Executive Summary:

This paper tries to explore the dynamic relationship between the yield of major crops in Bangladesh (aus, amas, boro, wheat and potato) and climate change (average rainfall, average maximum temperature and average minimum temperature). Graphical visualization shows that, there is an increasing trend in average annual minimum temperature and average annual minimum rainfall, while there is a decreasing trend of average annual maximum temperature. We applied the autoregressive distributive lag (ARDL) cointegration method in order to study the dynamic relationship. The paper finds long run dynamic link between the yield of aus rice and average rainfall and average minimum temperature. We have found short run dynamic relationship for a yield of aman rice, boro rice and potato. However, we do not find any dynamic relationship has been found for a yield of wheat with the variables of climate change. After estimation the dynamic relationship, we ensure there is no serial correlation among residuals using Breusch-Godfrey test and models are normally distributed using histogram and Jarque- Bera test. Finally, we ensure the stability of the models applying CUSUM test.

Keywords: Climate change, Food crops, Crops yield, Dynamic relation, Temperature, Rainfall,

Location: Bangladesh Bureau of Statistics

Start Date:

End Date:

Program Area: Risk Management in Agriculture

Non-commodity: Climate change

Objectives:

To find out the relationship of climate change with growth of yield of food crops (rice, wheat and potato). To overcome the above mentioned shortcomings of the literature, we will analyze the relation between climate change and food production system from another dimension.

Methodology:

We try to identify the long relationship between three major food crops (Rice, Wheat, Potato) and three important indicators of climate (Rainfall, Maximum Temperature, Minimum Temperature). As our research area is Bangladesh, annual data of from 1987 to 2015 are collected from Bangladesh Bureau of Statistics (BBS). To examine cointegration we apply Autoregressive Distributed Lag (ARDL) method. ARDL incorporate both lag value of regressand and regressors. It is a cointegration method when variables are stationary at both; level and first difference. To estimates the models, we use maximum lag 1 for both regress and regressor as we take only 28 years of data. Moreover, we take constant as trend specification. We consider the automatic selection process in order to select a model. After performing ARDL, we check the significance of the estimated model by the Bound test (Pesaran et al., 2001). If the test statistics of the bound test (F statistics and t statistics) are significant, the model will be considered as long run cointegrated. If the test statistics are insignificant, the model will be considered as not long run cointegrated. If the model is long run cointegrated, we estimate error correction model (ECM), where we check the speed of adjustment of both short run and long run to on another. Before checking the ECM, we identify lag length criteria following Schwarz information criterion (SIC). To do so, we adopt vector autoregressive (VAR) method. Later, we identify the residual series for the long run cointegrated model and ECM model. We use ordinary least square (OLS) method. For not significant long run cointegration, short run ARDL is estimated using OLS method. We check the diagnostic of residuals of estimated model. To study the normality, we observe the histogram and apply Jarque-Bera test. Moreover, to examines the autocorrelation we apply Breusch-Godfrey Serial Correlation LM test. In case of ECM and short run ARDL model we also check the stability of the model using the Cusum test.

Source of Information: International Journal of Sustainable Agricultural Research, 2020 Vol. 7, No. 4, pp. 287-303

Article Link (Online Journal): DOI: 10.18488/journal.70.2020.74.287.303

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

After completing estimation, yield of aus rice is long run cointegrated with one lag of average annual rainfall and average minimum temperature, which supports many empirical evidence, discussed earlier. Yield of aman rice, yield of boro rice and yield of potato are not long run cointegrate with the climate variables. But yield of aman rice has short run cointegration with one lag of average annual rainfall. Yield of boro rice has short run cointegration with the one lag of yield of boro rice. Yield of potato has short run cointegration with one lag of yield of potato and two lags of average annual rainfall. Yield of wheat does not have any cointegration. The result shows us that the impact of climate on crops production is still not much noticeable. In defense we assume that improved technology, soil fertility and fertilizer can be good reason behind this. However, the narrow impact of climate on production may suggest a greater magnitude impact in the long run.

Knowledge Management: Journal