Journal Volume: 68      No.: 3     Year: 2014
S.No Title Abstract Download
1 Fitting Exponential Smooth Transition Autoregressive Nonlinear Time-Series Model using Particle Swarm Optimization Technique
Author: Bishal Gurung, Ranjit Kumar and Himadri Ghosh      Pages: 327-332
Exponential Smooth Transition Autoregressive (ESTAR) family of parametric nonlinear time-series models is considered. The methodology for estimation of its parameters using a powerful Particle Swarm Optimization (PSO) technique is discussed. Further, simulation study is also carried out to test the validity of the proposed methodology. A heartening feature of ESTAR model is that, as opposed to some other nonlinear models involving regimes switching, the change between the extreme regimes is smooth and is assumed to be defined by a bounded continuous function of a transition variable. Further, it is capable of describing those datasets that depict cyclicity. As an illustration, it is employed for modelling and forecasting of Oil sardine, Mackerel and Bombay duck time-series landings data in India. Finally, the performance of fitted ESTAR model is also compared by computing goodness-of-fit statistic and various measures of forecast performance. It is concluded that fitted ESTAR model perform better than ARIMA methodology for the datasets under consideration. Keywords: Exponential Smooth Transition Autoregressive model, Particle Swarm Optimization, Cyclicity, Goodness-of-fit, Forecast performance, Fish landings data, ARIMA.
2 Estimation of Finite Population Total for Skewed Data
Author: Pradip Basak, Hukum Chandra and U.C. Sud      Pages: 333-341
In many surveys (for example, agriculture, business enterprises, income and expenditure surveys), data are typically skewed which contain few extreme values and linear model assumptions are questionable. Commonly used survey estimation methods for population total are based on normality assumption, that is, survey data are linear. As a consequence, these methods are both model biased and inefficient for skewed data. We describe estimation of finite population total for skewed data that are linear following a suitable transformation, in particular logarithmic transformation. We demonstrate the comparative performance of different estimators of population total for skewed data using both model based simulations as well as design based simulations. Empirical results clearly reveal that linear model based estimators are inefficient for skewed data. Keywords: Skewed data, Transformation, Bias correction, Model calibration, Prediction.
3 Stochastic Volatility Model Fitting using Particle Filter: An Application
Author: Bishal Gurung, Prajneshu and Himadri Ghosh      Pages: 343-350
In this article, we study the Stochastic volatility (SV) model in which volatility is an unobservable variable following some stochastic process. Procedure for estimation of parameters of this model using Particle Filter (PF), which is a powerful Sequential Monte Carlo technique, is discussed. To this end, relevant computer program is developed in MATLAB, Ver. 7.4 software package. As an illustration, All-India data of month-wise total export of Basmati rice during the period April, 2003 to June, 2013 is considered. Comparative study of the fitted SV model vis--vis Exponential Generalized autoregressive conditional heteroscedastic (EGARCH) model is carried out by computing various measures of goodness of fit. Subsequently, forecasting performances of SV and EGARCH models are also compared using several statistical measures. Finally, it is shown that SV model fitted through Particle filter performed better than EGARCH model for the data under consideration. Keywords: Exponential generalized autoregressive conditional heteroscedastic model, Heteroscedasticity, Particle filter, Stochastic volatility model.
4 Hindi Supplement
Author: ISAS      Pages: 429-433
5 Bayesian Predictive Inference for the Mean and Variance of a Finite Population Proportion: Two Stage Cluster Sampling with Non-Sampled Cluster Sizes Unknown
Author: Michael J. Racz and J. Sedransk      Pages: 351-358
With a complex survey design Bayesian predictive inference for finite population quantities may be difficult to carry out in organizations where skill in applying sampling-based methods is limited. Here we investigate the feasibility of approximating part of the analysis. Motivated by a survey of the quality of care that radiation therapy patients receive we assume a two-stage cluster sample design where the cluster sizes are known only for the clusters in the sample. We propose an exact analysis conditional on the cluster sizes for all units in the population, but an approximate analysis to take account of the unknown cluster sizes for the nonsampled clusters. Successful approximation will greatly simplify the analysis, and suggests the value of similar approximations when there are more complex sample designs. Keywords: Binary variable, Radiation therapy, Prostate cancer.
6 Addressing Disease Burdens Attributable to Ambient and Household Air Pollution in India: A Review to Scope Future Research Priorities for Carcinogenicity of Air Toxics
Author: Santu Ghosh, Kalpana Balakrishnan, Krishnendu Mukopadhyay, Sankar Sambandam, Naveen Puttaswamy, Moumita Chakraborty, Parthasarathi Ghosh, Manas Ranjan Ray, Dona Sinha and Saumyadipta Pyne      Pages: 391-405
Air pollution ranks among the leading risk factors contributing to the burden of disease in South Asia with household and ambient air pollution accounting for 6 percent and 3 percent respectively of the total national burden of disease in India. Both urban and rural communities bear this burden in terms of premature mortality and disability adjusted life years, resulting from excess risks of communicable and non-communicable diseases. We review the information pertaining to exposures to fine particulate matter and air toxics together with the attributable disease burden estimates. We also provide a summary of the results from recent assessments on carcinogenicity of ambient and household air pollution conducted by The International Agency for Research on Cancer. We conclude with a list of specific priorities for action related to air toxics and cancer in India. Keywords: Ambient air pollution, Household air pollution, Particulate matter, Air toxics, Disease burden, Carcinogenicity, Air quality actions.
7 District Level Food Insecurity in the State of Uttar Pradesh
Author: A.K. Nigam, R. Srivastava, P.P. Tiwari, Reeta Saxena and Shruti Shukla      Pages: 407-417
This paper critically examines the issues related to food insecurity at district level in the state of Uttar Pradesh in India. A total of 21 indicators were used to capture the three components of food insecurity viz; food availability, food access and food absorption. The methodology adopted was an improvement over the one adopted by M.S. Swaminathan Research Foundation (MSSRF). The modified approach used geometric mean instead of arithmetic mean used by MSSRF. Principal Component Analysis (PCA) was also tried in the present study though it was found to be useful only when indicators of components of food insecurity were dealt separately. Small area estimates were used wherever required. The districts were ranked on the basis of all the indicators. After ranking the districts, a mapping index was developed. In the present study, all the 70 districts have been put into five typologies namely ?extremely insecure', ?severely insecure', 'moderately insecure', ?moderately secure' and ?secure' based upon composite mapping index. Class intervals of each typology have been determined by GIS software Arc View, following natural breaks in the series. All the 70 districts of the state were categorized in these typologies and a food insecurity map was developed. Keywords: Food availability, Food access, Food absorption, Food insecurity, Principal Component Analysis, Small Area Estimates.
8 Nonlinear Support Vector Regression Methodology for Modeling and Prediction: An Application
Author: M.A. Iquebal, Prajenshu and Sarika      Pages: 359-364
The main limitation of Multiple linear regression analysis for estimating cause-effect relationship is highlighted. Artificial neural network (ANN) methodology that does not require specification of exact nonlinear functional relationship between a response and a set of predictor variables is briefly discussed. Some advantages and disadvantages of this technique are pointed out. The recently developed Nonlinear support vector regression (NLSVR) methodology, which is very promising and versatile, is described. As an illustration, Maize crop yield data as response variable and Total human labour, Farm power, Fertilizer consumption and Pesticide consumption as predictor variables are considered. Both ANN and NLSVR techniques for modelling and prediction purposes are employed. Performance of a fitted model is assessed in terms of Root mean square error (RMSE), Mean absolute error (MAE) and Mean absolute prediction error (MAPE). STATISTICA software package is used for carrying out data analysis. Superiority of NLSVR technique over ANN technique is showed for the data under consideration. It is concluded that NLSVR methodology is quite successful for modelling as well as prediction purposes. Keywords: Kernel function, Maize crop yield, Mean absolute prediction error, Multilayer perceptron, Nonlinear support vector regression, Polynomial, Radial basis function, Sigmoid.
9 Price Volatility in Agricultural Commodity Futures - An Application of GARCH Model
Author: R. Sendhil, Amit Kar, V.C. Mathur and Girish K. Jha      Pages: 365-375
Uncertain movement in commodity prices is a major concern for policy makers. Generalised autoregressive conditional heteroscedasticity (GARCH) model was applied to measure the extent of volatility in spot prices due to futures trading. The study sourced the available daily spot prices of selected twenty agricultural commodities that are traded in NCDEX platform both for 2009-10 (period of peak inflation) and right from the date of commencement of trading till June 2014. Empirical results indicated low price volatility in maize, soybean, cotton seed oilcake, castor, palm oil, cumin and chilli during the peak inflation period i.e., 2009-10; whereas, chickpea, cotton seed oilcake, mustard and cumin experienced the same level of volatility right from inception of trading. The present study concludes that futures market helps to reduce price volatility but not necessarily in all the commodities. Hence, it is recommended that the commodity exchanges should continue the trading in commodities that exhibit low volatility. Further, actual economic reasons for the persistence of volatility in the rest of the commodities have to be probed. Keywords: GARCH, Volatility, Futures trading, Agricultural commodity futures, NCDEX.
10 Efficient Row-Column Designs with Two Rows
Author: Sukanta Dash, Rajender Prasad and V.K. Gupta      Pages: 377-390
A general method of construction of efficient row-column designs with two rows has been developed. Lower bounds to A- and D- efficiencies of the designs obtained through the proposed method of construction for 3 ? v ? 10, v ? b ? v(v?1)/2, 11 ? v ? 25, b = v and (v, b) = (11, 13), (12, 14), (13, 14) and (13, 15) where v is the number of treatments and b is the number of columns have been computed and compared with those of best available designs in the above parametric range in the literature under a fixed effects model. The lower bounds to A- and D- efficiencies of the designs obtained have also been compared with those of best available designs in the literature under a mixed effects model (considering column effects as random). Robustness aspects of optimal/efficient designs has been investigated under a mixed effects model for different values of ? (a function of inter and intra column variances). Several designs have been obtained that are more efficient under mixed effects model as compared to the best available designs. A computer programme using C# programming language with ASP.NET platform has been developed for generating efficient row-column designs in two rows in the above restricted parametric range. The A- and D- efficiencies of the designs are also reported. Keywords: Row-column designs, A-efficiency, D-efficiency, Robustness.
11 Acknowledgement to Reviewers
Author: ISAS      Pages: 2
12 Online Progress Monitoring of Agricultural Scientists: e Initiative
Author: R.C. Goyal, Alka Arora, Sudeep Marwaha, P.K. Malhotra, Rajni B. Grover and A.K.M. Samimul Alam      Pages: 419-427
A web based system for Half-Yearly Progress Monitoring (HYPM) of the agricultural scientists working in ICAR institutes has been developed and hosted ( at Indian Agricultural Statistics Research Institute (IASRI), New Delhi. The system has been developed using three-tier web architecture on the ASP.NET technology platform. Authenticated secured access has been given to all concerned users; Scientists, Reporting Officers, Reviewing Officers, Nodal Officers and Research Managers involved in the monitoring process of the scientists. Nodal officer at each Institute is responsible for Institute level customization of HYPM and has the right to assign different roles for monitoring, issue password to scientific personnel and allocation of Scientists for reporting and reviewing under Reporting and Reviewing Officers of their respective institutes. Scientists have facility for online submitting their research targets and achievements under different heads of teaching, training, extension and other prioritized activities. Research Manager Personnel (RMP?s) of ICAR have the flexibility to view reports at all levels, i.e. Institutional, Subject Matter Division (SMD) and consolidated for the entire ICAR institutes on different parameters. Keywords: Agricultural Scientists, Half-Yearly Progress Monitoring, HYPM, ICAR.