Browsing by Author "Girish Kumar Jha"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
ThesisItem Open Access Agriculture Price Forecasting with Structural Break in Time Series Data(ICAR-Indian Agricultural Statistics Research Institute ICAR-Indian Agricultural Research Institute New Delhi-, 2020) RAJEEV RANJAN KUMAR; Girish Kumar JhaAccurate price forecasting of agricultural commodities is very important for raising income of the farmers as well as for avoiding market risk. However, due to biological nature of production of agricultural commodities, forecasting of their prices become a challenging task. These challenges become more severe when structural breaks are present in the observed agricultural price series due to factors like major changes in technology, sudden changes in economic policy, etc. In this study, an effort has been made to account for the structural break along with the other complex patterns like non-stationarity, non-linearity, long memory and cointegration present in the agricultural price series.. Generally, single model may not be able to capture all complex patterns present in the data series concurrently. Therefore, to capture various complex patterns in the data along with structural break, hybridization of statistical model that account for structural break with artificial intelligence model has been done. Accordingly, for agricultural price volatility forecasting in the presence of structural break, a hybrid model based on Markov-Switching GARCH (MS-GARCH) and Extreme Learning Machine (ELM) is proposed. The performance of the proposed hybrid MS-GARCH–ELM model is evaluated on the weekly potato price of Delhi market, monthly international Groundnut oil and Palm oil price series, and it is found that the proposed model outperformed its counterparts. Empirical results of agricultural price series that contain long memory property with structural break show that the forecasting performance of the proposed hybrid model based on ARFIMA with dummy variable combined with ELM is better than the individual model. Further, the effect of structural break in the co-integrated system has also been evaluated. Accordingly, spatial market integration among major Potato markets in India are investigated in the absence and presence of structural break. The overall co-integration test results indicated that selected potato markets in India are well integrated and have long-run price association across them.ThesisItem Open Access Co-integration approach for energy-use in agriculture(Indian Agricultural Statistics Research Institute Indian Agricultural Research Institute New DelhI, 2014) Rajeev Ranjan kumar; Girish Kumar JhaIn view of increasing share of energy in the cost of cultivation as well as deregulation of prices of some petroleum products, agricultural commodity prices are vulnerable to the rise in energy prices, particularly of crude oil. In this study an attempt has been made to examine the co-movement between energy and agricultural commodity prices with the help of Johansen cointegration technique using monthly wholesale price indices for the period April 1994 to March 2014. Since the process of deregulation started from April 2002 onward, the entire period was divided in two equal parts, so that before and after period analysis will provide a clear picture of a potential link between prices. The results clearly revealed that energy and selected agricultural commodity prices are integrated in the long-run since 2004 while fruits prices were integrated even before deregulation of petroleum price. This means that there is an increasing tendency for price changes in selected agricultural commodities corresponding to changes in international crude oil prices in recent years. Further an effort was also made to examine energy growth linkage in major states of India with the help of panel cointegration using annual time series data of real GSDP from agriculture and allied sectors and corresponding electricity consumption for agriculture during 1990-2010. The empirical analysis fully supported a positive long-run cointegrated relationship between GSDP and electricity consumption when the heterogeneous state effect was taken into account. It was observed that although agricultural growth and energy consumption lack short-run causality, there is long-run unidirectional causality running from energy consumption to agricultural growth. This implies that reducing energy consumption does not adversely affect agricultural growth in the short-run but would in the long-run, thus energy demand will increase in future in order to achieve higher agricultural growth. BThesisItem Open Access STUDY ON MACHINE LEARNING TECHNIQUES BASED HYBRID MODEL FOR FORECASTING IN AGRICULTURE(ICAR-Indian Agricultural Statistics Research Institute ICAR-Indian Agricultural Research Institute New Delhi, 2019) PANKAJ DAS; Girish Kumar JhaAgricultural datasets are mostly nonlinear, nonstationary and leptokurtic in nature. These properties of dataset pose a variety of problems in forecasting. Precise forecasting helps both farming community and policy makers to undertake informed decisions. Literature suggests that each of the forecasting models has their own limitations. A single forecasting model is not able to handle problems like nonstationary and nonlinearity simultaneously. Accordingly, the present study proposes three different hybrid models i.e. empirical mode decomposition based support vector regression (EMD-SVR), time-delay neural network with error correction term (TDNN-ECT) and multivariate adaptive regression splines based artificial neural network (MARS-ANN) models. The novelty of these models lies in the fact that they can handle both nonstationary and nonlinear features of dataset simultaneously. In EMD-SVR model, the nonstationary and nonlinear dataset is decomposed into different intrinsic mode functions and final residue through EMD method. Then the decomposed components are forecasted using SVR model and finally, all forecasted values are summed up to produce the final forecast. In the second model, TDNN-ECT uses the error correction term from the two co-integrated series as auxiliary variable. The auxiliary information in the form of ECT improves the forecasting accuracy. Further, selection of important input variables is a crucial step in determining the accuracy of any forecasting model. Hence, MARS-ANN hybrid model was developed in which the MARS algorithms was employed to extract important factors determining crop yield and the extracted factors were used for yield prediction using ANN methodology. The performance of proposed hybrid models is evaluated with individual forecasting models using three different agricultural datasets. The performance measures like RMSE, MAD, MAPE and ME are used to evaluate the model. The results indicated that the performance of the proposed hybrid models are substantially superior as compared to the individual forecasting model. Key words: Co-integration, Nonlinearity, Nonstationary, EMD, SVR, TDNN, and MARS.