Mahmuda Akter
Graduate School of Economics, Rissho University, Tokyo, Japan
Mizanur Rahman Sarker
Department of Agricultural Statistics, Sher-e-Bangla Agricultural University, Dhaka, Bangladesh
Climate change, Rice, Panel data, Temperature, Humidity, Rainfall, Bangladesh
Socio-economic and Policy
Climate change
Empirical Model Specification: Regression models (Sarkar et al., 2012; Boubacar, 2010; Mendelsohn, 2009; Isik and Devadoss, 2006; You et al., 2005; Peng et al., 2004) and indirect crop simulation models (Schlenker and Roberts, 2008) are used in most of the study on possible effects of climate variability and change on food crops. Also, Feasible Generalized Least Squares (FGLS) and Maximum Likelihood Estimation (MLE) can be used. However, FGLS estimation is employed in most empirical studies, although MLE is more efficient and unbiased than FGLS for small samples (Saha et al. 1997). Given the large sample size here, FGLS was used, as described in Judge et al. (1988), to estimate a form of fixed effects panel model. The panel data used in this study shows that it follows a normal distribution when histograms of rice mean yield against time were drawn. Non-climatic factors such as improved variety, management techniques, fertilizers, pesticides may cause changes in the mean yield of rice. Therefore, we need to remove the mean yield trend caused by non-climatic factors before we run our linear regression model. To remove the non-climatic trend and avoid heteroskedasticity in the linear regression model, we can use a log-linear regression model. Log-transformation can transform absolute differences into relative differences. Data Sources: Annual yield data of both local and high yielding varieties of three different rice crops (Aus, Aman, Boro) for the period of 2011-2018 were obtained from the Yearbook of Agricultural Statistics of Bangladesh (2017, 2015, 2013, 2011), Department of Agricultural Extension (DAE) and website of the Bangladesh Bureau of Statistics (BBS). These data were found according to the fiscal year, such as 2017-2018, 2018-2019, etc. Then, these fiscal year data were transformed into yearly data, for example, 2017-2018 was considered as 2018. Secondary data on monthly average temperature, monthly average total rainfall and humidity data from 2011 to 2018 has been collected from all 34 weather stations of the Bangladesh Meteorological Department (BMD) located all over Bangladesh. The three independent variables used for this study from this dataset are rainfall, temperature, and humidity. Rainfall is defined as the 12-month summations of monthly rainfall values. Temperature is the 12 month average of monthly average temperatures. Humidity is defined as the 12 month average of monthly average relative humidity. Testing for fixed versus random effects: In testing for fixed versus random effects, Hausman test were performed and random effect model was found to be appropriate (Appendix.IV). In light of the above results, panel corrected standard error (PCSE) estimates were obtained, which correct for cross sectional dependence, heteroscedasticity and autocorrelation. The parameters are estimated using a Prais Winsten (or OLS) regression. Equations have been estimated with district and year fixed effects. Regression was run for rice, explaining mean yield. Mean yield depends on climate and non-climate inputs. Our results, however, show mean yields are best explained by levels of rainfall and temperature. We surmise therefore it is variability in climate that makes agriculture more risky
The International Journal of Climate Change. Impact and Responses, volume-7, Issue-2
Journal