Journal Volume: 71      No.: 2     Year: 2017
S.No Title Abstract Download
1 A Non-parametric Regression based Computational Approach for Prediction of Donor Splice Sites
Author: Prabina Kumar Meher and A.R. Rao      Pages: 159-166
Identification of splice sites with higher accuracy is vital for systematic study of gene structures in eukaryotes. In this paper, an attempt has been made to develop a kernel regression based probabilistic approach for prediction of donor splice sites. The proposed method achieved an estimate of area under receiving operating characteristics curve of 93.750.56 and 93.500.56 and area under precision-recall curve of 96.130.43 and 96.130.43 on cattle (Bos Taurus) and fish (Danio rario) datasets respectively. The prediction accuracy of the developed approach was also found comparable with the existing probabilistic approaches viz., SAE, MEM, MDD, MM1 and WMM, while tested by using an independent splice site dataset. Thus, we believe that the proposed approach will supplement the existing approaches for prediction of donor splice sites. Keywords: Kernel regression, Indicator variable, Conditional expectation, Nucleotide dependencies.
2 Software for Design of Water Harvesting Ponds and Associated Structures
Author: Ramadhar Singh and Karan Singh      Pages: 177-186
Hydrologic design parameters (i.e. design rainfall, runoff and peak flow rate) database need to be developed on region basis for proper design of water harvesting structures in country?s on-going watershed development programmes at large scale. Software for the design of water harvesting pond and associated structures has been developed to make the designing task easy and simple. The software is developed using Visual Basic as front end and Microsoft Access as back end. The software has four basic modules ? hydrologic design, hydraulic design, structural design and estimation of materials and costs of water harvesting structures. Hydrologic design module consists of three sub-modules of rainfall frequency analysis for design rainfall, design runoff volume and design peak flow estimation. Hydraulic design module consists of five sub-modules for determination of storage capacity of pond based on Krimgold and Harold (1944) equation and crop water requirement, storage structure dimensions, spillway dimensions, and design of earthen embankment including seepage analysis. The structural design module includes two sub-modules of structures subjected to water pressure and earth pressures for checking the stability of water harvesting structures. The materials and costs estimation module consists of four sub modules related to different components. The software uses inbuilt database for different variables of formulae/equations used in runoff and peak flow estimation for ten identified districts of Madhya Pradesh (MP), India. The developed software was used to create database of design rainfall, runoff volume and peak flow and design of water harvesting structures in Vertisols. Keywords: Hydrologic design parameters, Frequency analysis, Database, Rainfall, Runoff .
3 Hindi Supplement
Author: ISAS      Pages: 5
4 Resolvable Multi-Session Sensory Designs Balanced for Carryover Effects
Author: Sumeet Saurav, Cini Varghese, Eldho Varghese and Seema Jaggi      Pages: 121-126
Sensory trials play a vital role in food and nutrition experiments in establishing certain sensory facts about agricultural/animal produce. To draw definite conclusion from the study, it is important to eliminate or minimize all sources of error and control all factors that may influence the inference. Hence, in addition to the potential sources associated with the preparation of the test products, variability due to measurement or assessment process, order effects, carryover effects and assessor fatigue are to be considered. An experimental design for sensory evaluation should be capable of accommodating all these variations. However, when there are a large number of products two operational constraints, viz. assessor constraint and preparation constraint, may limit the choice of experimental designs. Assessor constraint sets a maximum number of products that an assessor can evaluate within a session before onset of sensory fatigue and preparation constraint limits the number of products that can be prepared for a given session without loss of experimental control. Therefore, many times it may become necessary to split sensory evaluation into sessions. Here, a general method is developed based on initial sequences to construct designs for multi-session sensory trials balanced for carry over effects. In the proposed designs, all panelists will have to evaluate only a subset of samples in each session and they will not have to taste the same product more than once during different sessions. A possible way of analysis of data generated from such trials is also discussed. Keywords: Carry over effects, Initial sequence, Multi-session trials, Sensory trials, Variance balance, Williams Latin square.
5 Richards Stochastic Differential Equation Growth Model and its Application
Author: Himadri Ghosh and Prajneshu      Pages: 127-137
Richards four-parameter nonlinear growth model, which is a generalization of the well-known logistic and Gompertz models, is a very versatile model for describing many growth processes. However, one limitation of the corresponding Richards nonlinear statistical model is that it is applicable only when the data are available at equidistant epochs, which is not always possible. The other limitation is that it is not able to describe the underlying fluctuations of the system satisfactorily particularly for longitudinal data, as merely an error term is added to the deterministic model to obtain it. Accordingly, in this article, the general approach of ?Stochastic differential equations? is considered. Specifically, the methodology for Richards growth model in random environment is developed. The optimal predictor of untransformed data along with prediction error variance is also derived. Relevant computer programs for its application are written and the same are included as an Appendix. Finally, as an illustration, pig growth data are considered and superiority of our proposed model is shown over the Richards nonlinear statistical model for given data. Keywords: Richards nonlinear growth model, Stochastic differential equation, Interval estimation, Out-of-sample forecasting.
6 An Application of Fuzzy Programming Approach in Agriculture: A Case Study of Willow Wicker Cultivation in Kashmir
Author: M.A. Lone, S.A. Mir and Imran Khan      Pages: 139-146
This paper deals with decision making problems, where a fuzzy linear programming (FLP) approach with triangular and trapezoidal membership functions is used for optimal allocation of land for different crops such as rice, maize and willow wicker, locally known as Veer Kani in Kashmir, with respect to various factors. The solution of conversion of FLP into crisp multiobjective linear programming problems has been considered. Also, mean and median of triangular fuzzy numbers are considered to compare the results. Keywords: Fuzzy linear programming, Triangular membership function, Trapezoidal membership function, Maximizing income, Fuzzy set, Crop combination..
7 Model Based Calibration Approach for Estimating Population Total in Successive Sampling
Author: Nirupam Ghosh, U.C. Sud, Hukum Chandra and V.K. Gupta      Pages: 95-102
The conventional calibration approach is appropriate when study and auxiliary variables are linearly related. However, when study and auxiliary variables are non-linearly related model based calibration technique is appropriate. In this article two model based calibration estimators along with their variances and estimator of variances in two occasion successive sampling are proposed. The performance of the proposed estimators are studied via a simulation study vis--vis design based calibration estimator and an estimator which doesn?t consider auxiliary information at estimation stage. via a simulation study vis--vis design based calibration estimator and an estimator which doesn?t consider auxiliary information at estimation stage.
8 Modelling Repeated Measures with Multiple Endpoints: A New Approach
Author: Sandipan Samanta, Ranjit Kumar Paul and Arjun Roy      Pages: 113-120
In this article, the concept of O?Brien?s GLS test for repeated measure design has been introduced in heteroscedastic case. The paper is devoted to modify O?Brien?s test statistic and propose a methodology to analyze treatment effect using Kronecker product structure where multiple responses are recorded repeatedly over time. Additionally, unbalanced data can also be easily modeled with the proposed methodology. The proposed methodology is illustrated using orthopedic dataset. The size and power of the new test statistic is computed using bootstrap methodology for different parametric configurations. A comparison with doubly repeated measure design is also performed with existing methodology. It is concluded that the proposed methodology is better than Minimum Variance Quadratic Unbiased Estimation (MIVQUE) method and Restricted Maximum Likelihood (REML) procedure. Keywords: Bootstrap, Doubly repeated measure design, Heteroscedasticity, Kronecker product, O?Brien?s GLS test, Multiple responses.
9 An Improved ARFIMA Model using Maximum Overlap Discrete Wavelet Transform (MODWT) and ANN for Forecasting Agricultural Commodity Price
Author: Santosha Rathod, K.N. Singh, Ranjit K. Paul, Saroj K. Meher, G.C. Mishra, Bishal Gurung, Mrinmoy Ray and Kanchan Sinha      Pages: 103-111
Autoregressive fractionally integrated moving average (ARFIMA) is widely employed model for long memory time series forecasting in divergent domain from several decades. One of the main pitfall of this model is the presumption of linearity. As many long memory time series data in real world are not purely linear, therefore there is an opportunity to enhance the prediction ability of ARFIMA models by fusing with other nonlinear models. With this reasoning, the present article attempts to estimate the parameters of ARFIMA model by maximum overlap discrete wavelet transform (MODWT) and long memory time series prediction was made by combining ARFIMA-MODWT and ANN for forecasting spot prices of mustard. Experimental study justified the superiority of the proposed hybrid model over ARFIMA model in terms of several measurement indices. Keywords: ARFIMA, Long memory time series, MODWT, ANN, Hybrid methodology.
10 Comparative Performance of Imputation Methods for Different Proportions of Missing Data in Classification of Crop Genotypes
Author: Samarendra Das, Amrit Kumar Paul, S.D. Wahi and U.K. Pradhan      Pages: 147-153
Most crop datasets contain missing values, a fact which can cause severe problems in the analysis and limit the utility of resulting inference. Classification techniques for grouping of crop genotypes are used when the data is complete. However, the presence of missing values limits the utility of these techniques and creates bias in the resulting inferences. In majority of the cases, missing values are handled by deleting the genotype or traits which contain missing values there by losing information on these genotypes. An interesting approach to handle this problem is to impute the missing values. In this paper, we provided some solutions to handle missing data in crop breeding experiments for classification of crop genotypes. The performance of the imputation techniques is assessed by using the hit ratio criteria computed through four different classifiers by using extensive simulation procedure. This paper has also attempted to provide a description of missing data mechanism in agricultural experiments and various imputation techniques for missing data analysis in classification problems. For lower proportions of missing data, all four of the imputation techniques imputation techniques for missing data analysis in classification problems. For lower proportions of missing data, all four of the imputation techniques provided satisfactory results for classification of crop genotypes. For moderate level of missingness in the data, regression and multiple imputation techniques provided same levels of precision for classification of crop genotypes. When there is a high proportion of missing data, multiple imputation technique outperformed all imputation techniques for classification of crop genotypes. Among the classifiers, k-th nearest neighbor is the best classification technique in missing data situations. Keywords: Missing values, Genotypes, Classification, Mean imputation, Regression imputation, Multiple imputation, Hit ratio.
11 Modelling of Population Growth for a Seasonal Incidence of Mustard Aphid, Lipaphis erysimi
Author: N. Okendro Singh, N. Gopimohan Singh, A.K. Paul, Surinder Kumar and Gangmei Sobha      Pages: 155-157
In the present study a method of fitting a nonlinear growth model to the data of mustard aphid population is discussed. It shows the appropriateness of the aphid population growth model to the data set. The highest aphid population is predicted in the first standard meteorological week by the fitted model. Keywords: Nonlinear, Growth model, Population, Productivity, Pest.
12 Development of GIS based Decision Support System for Estimation of Watershed Surface Runoff
Author: P.D. Sreekanth, K.V. Kumar, S.K. Soam, N.H. Rao and A. Krishna Prasad      Pages: 167-175
Runoff from rainfall is an important component of the hydrological cycle. Estimation of runoff is critical for the design of hydrological structures and drainage systems in watersheds. Different soils, land use and water management practices affect runoff differently. In real watersheds, land use, soils and weather conditions vary spatially over the geographical area of the watershed leading to spatial variations in runoff. Also, most watersheds form a part of larger drainage basin or a large watershed consisting of several such sub-watersheds. Each sub-watershed is hydrologically connected to the other sub-watersheds of the basin. A comprehensive Geographical Information System (GIS) based decision support system (DSS) for estimation of runoff that includes the spatial variations in rainfall and natural resources is presented. A DSS is a computer-based information system which serve the management, operations, and planning levels of an organization and help people make decisions about problems that may be rapidly changing and not easily specified in advance. The DSS is developed as a deployable application by integrating independent GIS layers of watershed features created in ArcGIS with MapObjects in Visual Basic software, as a case study for the KK3 watershed in Mahaboobnagar district of Telangana state, India. The DSS generates thematic maps of spatial variations in runoff on individual rainy days for the sub-watersheds. Keywords: GIS based DSS, GUI, Land use, Rainfall, Remote sensing, Runoff, Soil, Sub-watershed, Watershed.