On the first order regression procedure of estimation for incomplete regression models

This article discusses some properties of the first order regression method for imputation of missing values on an explanatory variable in linear regression model and presents an estimation strategy based on hypothesis testing. This work was carried out before Professor V.K. Srivastava passed away in 2001. The author is grateful to the referees for their illuminating comments on an earlier draft of this paper.     

Educational Psychology in India: Its Present Status and Future Concerns

This study examined the status of educational psychology in the general curricula of psychology and education at various academic levels in India. It also focused on the published research in this area during the last 15 years. Analysis reveals that educational psychology in India is struggling for a respectable place in the psychology curricula. Despite the fact that it has been included as a foundation course in education curricula, the course of study is characterized by an arbitrary selection of topics and stereotypic, inappropriate, outdated contents. Fewer research studies are conducted in educational psychology than in some other branches of psychology in India. A lack of in-depth analysis of the research problems and of collaborative efforts, a failure to keep up with recent international research and theorizing, an excessive use of the questionnaire method with easily available respondents, and instrument-dependent research are some of the characteristics of educational psychology research in India. To make educational psychology more responsive to the changing social needs, suggestions are made regarding changing the orientation and emphasis in the teaching of the discipline, and altering research priorities within the framework of the country's historical and sociocultural contexts.     

Interval estimation of the largest reliability of k weibull populations

A procedure is given for obtaining a random width confidence interval for the largest reliability of k Weibull populations. The procedure does not identify the populations for which the reliability would be a maximum. The maximum likelihood estimators or the simplified linear estimators of the reliability based on type II censored data are used. The cases considered include unknown shape parameters being equal or unequal. Simultaneous confidence intervals for the k reliabilities are also obtained. Tables for the lower and upper limits in selected cases are constructed using Monte Carlo methods.     

Stein-rule estimation in models with a lagged-dependemt variable

There is an extensive literature on the Stein-rule estimation of the parameters in a regression model. An important result in this literature is that the Stein-rule does not dominate the least squares estimator in the lower mean squared error sense when there are two or less regressors in the model. However, we note that not much is known about the Stein-rule estimation in dynamic models. This paper is a modest attempt in this direction and it shows that Stein-rule estimation of the lagged coefficient does not dominate the least square indicating that the result of regresion model goes through in the dynamic case.     

Tests for exponentiality against new better than old in expectation and new better than some used in expectation alternatives

We present statistical procedures for testing exponentiality againt New Better than Old in Expectation (NBOE) and New Better than Some Used in Expectation (NBSUE) alternatives. The test statistics devised for the purpose are U-Statistics and hence asymptotically normally distributed. Pitman's asymptotic relative efficiency results have been obtained and Monte Carlo study presented to compare power of the test proposed with the other tests.     

Pattern and Determinants of Paternal Involvement in Childcare: An Empirical Investigation in a Metropolis of India

An analysis of various aspects of paternal involvement among Indian fathers is presented in this article. The pattern of involvement in terms of the activities participated in, and their frequency of participation have been examined. Overall level of involvement of fathers in childcare has also been determined. A number of hypotheses regarding predictors of paternal involvement have been formulated and examined with the use of empirical data. The study is based on interviews conducted with Indian couples (N = 350) having at least one child aged 10 years or younger. To understand the determinants of paternal involvement, regression has been conducted wherein paternal involvement has been regressed with 14 predictors. Results indicate that paternal involvement in childcare is mainly determined by the perception of individuals towards fatherhood—be it gender role expectations, or perception of the peer group, or fathering received by the individual fathers. Other socio-economic factors affecting paternal involvement have also been examined. As the importance of paternal involvement in childcare is increasingly being recognised, these findings have implications for programmers and policy makers. Interventions to bring about a positive change in the attitude of fathers may improve paternal participation.     

On the choice of transformations of the correlation coefficient with or without an outlier

Five transformations of the correlation coefficient, namely, Fisher's z, Nair's u, Sankaran's v, Ruben's y and Samiuddin's t are compared numerically using confidence intervals. Samiuddin's t_s transformation is close to the exact nominal confidence level for a small sample size ≤ 25 from a bivariate normal density. For a sample size > 25 both Samiuddin's t_s and Fisher's z can be used. In the presence of an outlier (on a minor axis), both Fisher's z and Samiuddin's t_s are not affected as long as |p| ≤ 0.3 but are seriously affected when |p&| > 0.3.     

On the distribution of hotelling's t2 and multiple correlation r2when sampling from a mixture of two normals

In this paper the non-null distribution of Hotelling's T² and the null distribution of multiple correlation R² are derived when the sample is taken from a mixture of two p-component multivariate normal distributions with mean vectors μ₁ and μ₂ respectively and common covariance matrix ∑, ∑. In a special case the non-null distribution of R² is a l s o given, while the general noncentral distribution is given i n Awan (1981). These results have been used to study the robustness of T² and R² tests by Srivastava and Awan (1982), and Awan and Srivastava (1982) respectively.     

On selecting populations better than a control: weibull populations case

Consider k( k ≥ 1) independent Weibull populations and a control population which is also Weibull. The problem of identifying which of these k populations are better than the control using shape parameter as a criterion is considered. We allow the possibility of making at most m(0 ≤ m < k) incorrect identifications of better populations. This allowance results in significant savings in sample size. Procedures based on simple linear unbiased estimators of the reciprocal of the shape parameters of these populations are proposed. These procedures can be used for both complete and Type II-censored samples. A related problem of confidence intervals for the ratio of ordered shape parameters is also considered. Monte Carlo simulations as well as both chi-square and normal approximations to the solutions are obtained.