Journal Volume: 66      No.: 3     Year: 2012
S.No Title Abstract Download
1 Estimation and Prediction under Nonignorable Nonresponse via Response and Nonresponse Distributions
Author: Abdulhakeem A.H. Eideh      Pages: 359-380
The response distribution is the distribution of the observed outcomes given the respondent set and sample units. We study the response and nonresponse distributions under nonignorable nonresponse. We give some new results that further favor the use of the response and nonresponse distributions for analytical inference of complex survey data under nonignorable nonresponse. We derive some new relationships between moments of the population distribution before sampling and the response and nonresponse distributions. Thus provides new justification for the broad use of probability-weighted estimators (design- based school) in estimating finite population parameters in case of ignorable nonresponse. In addition to the estimation problem we introduce new predictors of the finite population total, under common mean population model, simple ratio population model, and simple regression population model. These new predictors take into account the nonignorable nonresponse. Thus, also provides new justification for the broad use of best linear unbiased predictors (model-based school) in predicting finite population parameters in case of ignorable nonresponse. The main feature of the present estimators and predictors is their behaviours in terms of the nonignorable nonresponse parameters. Furthermore, we introduce two new tests for testing the ignorability of nonresponse. Keywords: Nonignorable nonresponse, Response propensity, Response distribution.
2 Studies on Some Preliminary Test Estimators in Double Sampling
Author: Phrangstone Khongji and Gitasree Das      Pages: 381-390
It is known that in many of the large scale surveys, it is inevitable to adopt stratification for the purpose of preparing a frame from which the sample can be extracted. Cochran (1977) suggested a regression estimate in stratified sampling which he called a combined regression estimate. In the present study, situations will be considered where partial information about the mean of the auxiliary variable is available. In order to utilize the partial information, double sampling is used and a preliminary test is done to construct the combined regression preliminary test estimator. The bias, mean square error and relative efficiency are obtained for the suggested estimator. Apart from analytical results, these are also obtained by numerical techniques. The comparative study shows that the bias and mean square error function obtained by numerical methods depict similar pattern with that obtained by analytical methods. In order to judge the performance of the suggested estimator, besides analytical results, empirical work is also carried out with the help of both real life data as well as simulated data. Recommendation of the levels of the preliminary test and optimum allocation of sample sizes are given. Keywords: Double sampling, Preliminary test estimator, Regression estimator.
3 On Small Area Estimation Techniques - An Application in Agriculture
Author: B.V.S. Sisodia and Anupam Singh      Pages: 391-400
Small area estimation approach development by Sisodia and Singh (2001) is revisited. This approach does not require any additional survey or conducting extra CCEs for crop-production estimate at block level. The district level data on crop- production and related auxiliary variables are exploited through regression models to obtain reliable estimate of crop production at block level. A new scaled estimator of block estimate is also proposed. Some alternative procedures to obtain weights for the auxiliary variables based on partial correlation and standardized regression coefficients are also suggested. An empirical study for wheat production in Barabanki district of State Uttar Pradesh (India) suggests that among the scaled estimators, Y?(2)q is the best one and the best choice of weights (wj) be based on partitioning of sum of squares due to regression. Some limitations of the study are also highlighted. Keywords: Small area statistics, Scaled estimators, Regression model, Crop cutting experiments.
4 Optimum Stratification for Sensitive Quantitative Variables using Auxiliary Information
Author: Med Ram Verma, Sarjinder Singh and Rajiv Pandey      Pages: 401-412
The paper considers the problem of optimum stratification for two sensitive quantitative variables when data on sensitive variables are collected by scrambled randomized response technique and an auxiliary variable is taken as stratification variable. We have proposed a cumulative cube root rule for determination of optimum strata boundaries for ratio and regression method of estimation under compromise method of allocation. A limiting expression for the trace of variance covariance matrix and an approximate expression for sample size also have been suggested. The paper concludes with numerical illustration. Keywords: Sensitive variable, Scrambled response, Optimum stratification.
5 Application of Bayesian Elastic Net and Other Shrinkage Methods in Genomic Selection and QTL Mapping
Author: Paulino Perez, Gustavo de los Campos, Sussane Dreisigacker, Hector Sanchez-Villeda and Jose Crossa      Pages: 413-426
The Elastic Net is a variable selection and shrinkage estimation method especially designed for regression settings with a large number of correlated predictors. Recently, a Bayesian formulation of the Elastic Net was proposed (BEN=Bayesian Elastic Net). In this article, we extend the BEN to model the combined effects of dense molecular markers and pedigree data and evaluate the performance of the proposed model using a barley data set and two large wheat data sets. The predictive power of the proposed model was compared with those of two well-established models: the Bayesian LASSO and the Bayesian Ridge Regression. Results show that the prediction assessment of BEN was as accurate as those of the other methods in all studied cases. The number of molecular markers with significant effects detected by BEN in four data sets was compared with those found by the Bayesian LASSO and Bayesian Ridge Regression models. An R-program that implements the proposed model is available. Keywords: Bayesian Elastic Net, Shrinkage methods, Genomic selection.
6 Estimation on Finite Population Variance using Partial Jackknifing
Author: Sarjinder Singh and Inderjit Singh Grewal      Pages: 427-440
In this paper, a new idea of partial jackknifing to estimate the variance of the ratio type estimator of the finite population variance due to Isaki (1983) in the presence of random non-response has been introduced. The proposed estimator has been compared with three different estimators of the variance through an empirical study. Keywords: Estimation of variance, Jackknifing, Auxiliary information.
7 D-optimal Designs for an Additive Quadratic Mixture Model with Random Regression Coefficients
Author: Manisha Pal and Nripes Kumar Mandal      Pages: 441-446
In a mixture experiment, the mean response is assumed to depend only on the relative proportion of ingredients or components present in the mixture. Scheffé (1958, 1963) first systematically considered this problem and introduced different models and designs suitable in such situations. The problem of estimating parameters in a mixture model has been considered by many authors. However, in their studies, they assumed fixed regression coefficient models. In this paper, we consider an additive quadratic mixture model with random regression coefficients and find the optimum design for the estimation of mean regression coefficients using D- optimality criterion. Keywords: Mixture experiments, Additive quadratic model, Random regression coefficients, Barycentres, D-optimality criterion.
8 Two Stage Sampling for Estimation of Population Mean with Sub-sampling of Non-respondents
Author: U.C. Sud, Kasutav Aditya, Hukum Chandra and Rajender Prasad      Pages: 447-457
The estimation of population mean in the presence of non-response has been considered when the sampling design is two-stage. Considering, three different cases of nonresponse, the corresponding estimators based on sub-sampling of non- respondents, collecting data on the sub-sample through specialized efforts, are developed. Expressions for the variances of the estimators along with unbiased variance estimators are developed. Optimum values of sample sizes are obtained by considering a suitable cost function. The percentage reduction in the expected cost of the proposed estimators are studied empirically. Keywords: Cost function, Non-response, Population mean, Sub-sampling, Two-stage sampling, Percentage reduction in the expected cost.
9 Building and Querying Soil Ontology
Author: Manoranjan Das, P.K. Malhotra, Sudeep Marwaha and R.N. Pandey      Pages: 459-464
Soil Taxonomy is based on soil properties that can be objectively observed and measured. There are many soil classification systems but USDA Soil Taxonomy is most accepted worldwide. Ontologies are the new form of knowledge representation that act in synergy with agents and Semantic Web Architecture. Ontologies define domain concepts and the relationships between them, and thus provide a domain language that is meaningful to both human beings and computing machines. The relationships in Ontology are explicitly named and developed with specification of rules and constraints so that they reflect the context of domain for which the knowledge is modeled. Ontologies can be built by using various GUI based software tools, known as Ontology editors. Among all editors Protégé [Gennari et al. (2003); Golbeck et al. (2003)] is widely supported by a huge research community. For effective use of Ontology, Protégé provides a query interface known as SPARQL query panel. SPARQL is a syntactically-SQL-like language for querying RDF graphs [Clark (2008)]. Soil ontology developed for USDA soil taxonomy will be helpful for study of soil taxonomy and classification of new soils. Soil Ontology is built in the Protégé OWL editor from Order to Sub group level. Using this soil ontology, a query interface can be developed that will help in detailed study of soil taxonomy, classification of new soil as well as exchange knowledge between software agents and systems. Keywords: RDF, OWL, Protégé, SPARQL, Ontology.
10 Acknowledgement to Reviewers
Author: ISAS      Pages: 465-470
11 Hindi Supplement
Author: ISAS      Pages: 5
12 Other Publications
Author: ISAS      Pages: 1