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

Agricultural and Fertilizers Production in Bangladesh: an Analysis of Their Interrelationships

Md. Sharif Hossain Ph. D.
Assistant Professor
Department of Accounting and Information Systems, University of Dhaka.

Md. Kamal Uddin
Associate Professor
Department of International Business, University of Dhaka.

Abstract/ Executive Summary:

The principal purpose of this paper has been made an empirical investigation to find whether the production of fertilizer causes the agricultural development or vise versa in Bangladesh. Apart from this is another purpose to investigate the existence of the long-run consequence in the series of agricultural production and production of fertilizers by using modern econometric technique. Specially here we used the unit root tests to know the existence of non-stationarity in the series of agricultural production, and production of fertilizers and the co-integrated tests is applied to know whether these variables share the same stochastic trend or not. In this paper, the Granger causality test is applied to clarify the causal relationship between the agricultural production and the production of fertilizers. Using regression analysis we have also examined the statistical association between the agricultural production and production of fertilizers. From the ADF and PP tests results, it has been found that hypothesis of stationarity has been uniformly rejected for these variables with and without structural breaks. Thus, it can be concluded that any innovation of these series will be permanent and have a long-run effect. From the Engle-Granger and Phillips and Ouliaris tests results it has been found that there is no long-run equilibrium relationship between these pairs of variables. From the Granger Causality test results it has been found that production of fertilizers causes the agricultural development which is significant at 10% level of significance but at 5% level of significance is not significant but the agricultural production does not cause the production of fertilizers From the estimated results of regression analysis it has been found that the production of fertilizers has positive effect on agricultural production but is not statistically significant.

Keywords: Long-run Consequences, Unit Root and Co-integration Tests, Granger Causality Test, Stationarity, Regression Analysis

Location: In Bangladesh

Start Date:

End Date:

Program Area: Farming System

Commodity: Agriculture, Fertilizer

Objectives:

To investigate whether or not our non-stationary result is sensitive to structural breaks. Instead of choosing a breakpoint arbitrarily on pre-test examination of the data, here the Zivot-Andrews (1992) procedure has been applied to search endogenously for the breakpoint and test for a unit root in each of the two variables.

Methodology:

For empirical investigation to find the association between agricultural production and production of fertilizers I have to collect a set of data. Data have been obtained on variables, the agricultural production (in thousand metric tons), and production of fertilizers (in thousand metric tons). It is obviously true that the success of any statistical and econometric analysis ultimately depends on the availability of appropriate data. It is therefore essential that we should spend some times discussing the nature, sources and limitations of the data that may arise in empirical analysis. The difficulty lies in the availability and nature of the data. A particular problem facing the researcher is to obtain the appropriate data. In Bangladesh a reliable data is a golden deer. It is often difficult to obtain good reliable data with the necessary information required for a particular analysis. Missing value is a great problem in some data. Even in some experimentally collected data, errors of measurement arise from approximation and rounding off. Because of all these and many other problems the researchers should always keep in mind that the result of research may be affected by the quality of the data. Therefore, if in given situations the researchers find that the results of research are unsatisfactory the case may be not they used the wrong model but the quality of the data was poor. Some difficulties in finding suitable data are frequently encountered. Keeping in mind different types of problems, for this study a data set is collected from Bangladesh Yearbooks of agricultural Statistics and Bangladesh Statistical Yearbooks corresponding to different years. This study covers available data from 1971-72 to 2003-04. The principal econometric analysis involves testing whether the agricultural production and production of fertilizers contain a unit root or not and whether an underlying co-integration relation exists. If both the series does not contain a unit root then any innovation to the series are temporary and have short-run consequences. Otherwise, any shocks to it will be permanent and thus have a long-run effect. To test for the stationarity of a variable a number of econometric techniques have been developed which are called unit root tests. Next, I will move for unit root test. It is well known that the usual techniques of regression analysis can result in highly misleading conclusions when variables contain stochastic trends (Stock and Watson (1988), Nelson and Kang (1981), Granger and Newbold (1974),). In particular, if the dependent variable and at least one independent variable contain stochastic trends, and if they are not co-integrated, the regression results are spurious, (Phillips (1986), Granger and Newbold (1974)). To identify the correct specification of the model, an investigation of the presence of stochastic trend in the variables is needed. To test for the stationarity of the agricultural production and production of fertilizers a number of econometric techniques have been employed. The first one is the Augmented Dickey-Fuller (ADF) test and the second one is the Phillips-Perron (PP) (1988) test. Lettx is the series under investigation of unit root and stands for the first difference and the lagged difference terms on the right-hand side of the equations.

Source of Information: Social Science Review, The Dhaka University Studies, Part-D, Vol. 25, No. 2, December 2008

Article Link (Online Journal):

Funding Source:

Total Budget (Tk.) :

Achievement/Findings:

The test results support that the production of fertilizers causes agricultural production which is significant at 10% level but at 5% level of significance the causal relationship is not statistically significant, the agricultural production does not causes the production of fertilizers. In this paper I have also conducted regression analysis for investigation on the economic association between the change of agricultural production and change of production of fertilizers. From the estimated results it has been found that the 30 production of fertilizers has a positive effect on agricultural production. From the t-test and F-test results, we see that the effects of the variable on agricultural production are not statistically significant at 5% level of significance but statistically significant at 10% level of significance. From the Jarque-Bera (JB) test it has been found that the residuals are normally distributed with mean zero. From the estimated results of the Lagrange Multiplied test (LM) it has been found that these residuals are not auto-correlated and from ARCH test they it can be said that these residuals are not conditionally heteroscedastic.

Knowledge Management: Journal