Nonparametric Bayes and empirical Bayes estimation of the population median,with application in finite-population sampling

ABSTRACT

Nonparametric Bayes and empirical Bayes estimators of the population median are provided under Dirichlet process priors. The finite-population sampling is used to estimate the finite-population median under Dirichlet process priors. The asymptotic properties of the estimators are obtained from a frequentist perspective.     

Spatio-Temporal Modeling of Residential Sales Data

This article focuses on the location, time, and spatio-temporal components associated with suitably aggregated data to improve prediction of individual asset values. Such effects are introduced in the context of hierarchical models, which we find more natural than attempting to model covariance structure. Indeed, our cross-sectional database, a sample of 7,936 transactions for 49 subdivisions over a 10-year period in Baton Rouge, Louisiana, precludes covariance modeling. A wide range of models arises, each fitted using sampling-based methods because likelihood-based fitting may not be possible. Choosing among an array of nonnested models is carried out using a posterior predictive criterion. In addition, one year of data is held out for model validation. A thorough analysis of the data incorporating all of the aforementioned issues is presented.     

Higher Education in Basic Science and Socioeconomic Characteristics of Students’ Life in India: An Exploratory Study

A shortage of scientifically trained manpower in India today is largely due to the fact that fewer and fewer students are being attracted to pursue higher studies in basic science, as current science enrolment data reveal. Ultimately, because of this shortage, developments in science and technology and science-based industry are slowing down, and the basic qualities of material life are continually being compromised. Several other countries are also contending with this problem, but India’s case has largely remained underrepresented in current research. To improve the situation, a number of competitive scholarship programs have been started in India to attract young students to pursue higher studies in science in greater numbers. For one such program, using a sample of more than 8,600 high-school students, we examine how a number of socioeconomic factors govern students’ decision to pursue science higher studies in the country. We find that student age, gender, family income, exam scores, use of computers and the internet, travels, and possession of modern consumer goods at home contribute to students’ increased mobility in search of better educational facilities needed for the preparation of a career in science. Our results have implications for schools and teachers, globalization, scientific literacy, and the general prosperity of life in the country.     

Growth,inequality and innovation: A CGE analysis of India

This paper probes into the growth and distributional consequences of four basic policy options emanating from the three sources of economic growth, namely, physical capital, human capital and technological progress, with the help of a computable general equilibrium model of India. The simulation results show that, the efficacy of physical capital accumulation in augmenting growth and abating income inequality is greater than that of human capital accumulation. In the long term, however, the latter overtakes the former in promoting growth, but inequality worsens. When the two policies are commingled, growth improves but it continues to be inequality-augmenting. Finally, with concomitant Hicks-neutral technological progress, not only is growth enhanced further, but it turns out to be significantly inequality-mitigating. The emerging policy lesson is that any integrated policy of boosting investments in physical as well as human capital must be closely bound up with technological progress for growth to be inclusive.     

Information in an observation in robust designs

We consider the Information contained 1n each observation in a given design robust with respect to the estlmability of parameters and against the unavailability of observations. We compare the observations in various 1-, 2- and 3- dimensional designs on the basis of their informations.     

Shewhart x-charts with estimated process variance

Properties of the Shewhart X-chart for controlling the mean of a process with a normal distribution are investigated for the situation where the process variance Ó²must be estimated from initial sample data. The control limits of the X-chart depend on the estimate of Ó²and thus, unlike the case when Ó²is known, the X-chart is not equivalent to a sequence of independent tests. When Ó²is estimated the distribution of the run length is not geometric and cannot be characterized simply in terms of the probability of a signal at a given point. The average run length (ARL) for the X-chart is expressed in terms of an integral involving the normal cdf, and it is shown that the chart signals with

probability one, but the ARL may not be finite if the size of the 2 sample used to estimate Ó²is sufficiently small. In addition, certain bounds for the ARL are also derived. Numerical integration is use to show that the effect of using small sample sizes in estimating Ó²is to increase the ARL and the variance of the run length distribution     

Optimal Retail Return Policies with Wardrobing

Many retailers offer refunds to consumers who, after a trial period, return a product that they find does not fit their needs. Some consumers are willing to use this return option opportunistically for short‐term consumption rather than its intended purpose of resolving fit uncertainty. Such behavior has been termed “wardrobing.” Restocking fees (partial refunds) can be used to combat wardrobing. However, there is a trade‐off involved, since partial refunds will be viewed negatively by consumers who return an item due to a true lack of fit. In this study, we consider how the extent of wardrobing (how many consumers consider such behavior) and the benefit of wardrobing (how much value can be extracted during the trial) impact firm pricing decisions and profits in this retail context. Our results imply that an increase in the extent of wardrobing is most detrimental to profits when the current extent of wardrobing is low. On the contrary, if the extent of wardrobing is already very high, and the benefit of wardrobing to consumers is also high, the retailer can set prices and refunds such that additional wardrobing actually increases firm profits. In a model extension, we show how a retailer can effectively screen wardrobers from ordinary consumers by offering a menu of price/refund pairs, and that such an approach can lead to increased profits if the extent of wardrobing is sufficiently high. Overall, our findings provide new insights into how retailers can set prices and refund policies to effectively manage opportunistic behavior by consumers.     

Robustness of complete diallel crosses plans to the unavailability of one block

The present investigation involved the estimation of the general combining ability of CDC plans subject to the unavailability of one block for Griffing's system IV. Further, it has been shown that CDC plans are fairly robust to the unavailability of one block.