Overdispersed negative binomial regression models

Overdispersion is a common phenomenon in actual data sets. It is important to have methods of dealing with extra variation in regression situations. This article develops tests for extra-negative binomial variation and gives some numerical methods to deal with it. Simulations show power comparisons between some of the overdispersion tests discussed. Finally, an example is given to demonstrate the methods.     

Large Differences between Partial Correlation Coefficients

A plot of each ranking of N objects in N-dimensional space is shown to provide geometric interpretations of Kendall's tau and Spearman's rho and also of the relationship of rho to a sum of inversion weights. The computation of rho from a sum of inversion weights is shown to allow sequential calculation of rho.     

A moment-generating function with application to the birnbaum-saunders distribution

The Birnbaum-Saunders distribution is a fatigue life distribution that was derived from a model assuming that failure is due to the development and growth of a dominant crack. This distribution has been shown to be applicable not only for fatigue analysis but also in other areas of engineering science. Because of its increasing use, it would be desirable to obtain expressions for the expected value of different powers of this distribution.

In this article, the moment-generating function for the sinh-normal distribution is derived. It is shown that this moment-generating function can be used to obtain both integer and fractional moments for the Birnbaum-Saunders distribution. Thus it is now possible to obtain an expression for the expected value of the square root of a Birnbaum-Saunders random variable. A general expression for integer noncentral moments for the Birnbaum-Saunders distribution is derived using the moment-generating function of the sinh-normal distribution. Also included is an approximation of the moment-generating function that can be used fcx small values of the shape parameter.     

Conditional moving linear regression: Modeling the recruitment process for ALLHAT

Effective recruitment is a prerequisite for successful execution of a clinical trial. ALLHAT, a large hypertension treatment trial (N = 42,418), provided an opportunity to evaluate adaptive modeling of recruitment processes using conditional moving linear regression. Our statistical modeling of recruitment, comparing Brownian and fractional Brownian motion, indicates that fractional Brownian motion combined with moving linear regression is better than classic Brownian motion in terms of higher conditional probability of achieving a global recruitment goal in 4-week ahead projections. Further research is needed to evaluate how recruitment modeling can assist clinical trialists in planning and executing clinical trials.     

Bayesian misclassification and propensity score methods for clustered observational studies

Bayesian propensity score regression analysis with misclassified binary responses is proposed to analyse clustered observational data. This approach utilizes multilevel models and corrects for misclassification in the responses. Using the deviance information criterion (DIC), the performance of the approach is compared with approaches without correcting for misclassification, multilevel structure specification, or both in the study of the impact of female employment on the likelihood of physical violence. The smallest DIC confirms that our proposed model best fits the data. We conclude that female employment has an insignificant impact on the likelihood of physical spousal violence towards women. In addition, a simulation study confirms that the proposed approach performed best in terms of bias and coverage rate. Ignoring misclassification in response or multilevel structure of data would yield biased estimation of the exposure effect.     

A Seemingly Unrelated Nonparametric Additive Model with Autoregressive Errors

This article considers a nonparametric additive seemingly unrelated regression model with autoregressive errors, and develops estimation and inference procedures for this model. Our proposed method first estimates the unknown functions by combining polynomial spline series approximations with least squares, and then uses the fitted residuals together with the smoothly clipped absolute deviation (SCAD) penalty to identify the error structure and estimate the unknown autoregressive coefficients. Based on the polynomial spline series estimator and the fitted error structure, a two-stage local polynomial improved estimator for the unknown functions of the mean is further developed. Our procedure applies a prewhitening transformation of the dependent variable, and also takes into account the contemporaneous correlations across equations. We show that the resulting estimator possesses an oracle property, and is asymptotically more efficient than estimators that neglect the autocorrelation and/or contemporaneous correlations of errors. We investigate the small sample properties of the proposed procedure in a simulation study.     

Maximum likelihood estimation in a Weibull regression model with type-1 censoring: a Monte Carlo study

Results of the Monte Carlo study of the performance of a maximum likelihood estimation in a Weibull parametric regression model with two explanatory variables are presented. One simulation run contained 1000 samples censored on the average by the amount of 0-30%. Each simulatedsample was generated in a form of two-factor two-level balanced experiment. The confidence intervals were computed using the large-sample normal approximation via the matrix of observed information. For small sample sizes the estimates of the scale parameter b of the loglifetime were significantly negatively biased, which resulted in a poor quality of confidence intervals for b and the low-level quantiles. All estimators improved their quality when the nominal value of b decreased. A moderate amount of censoring improved the quality of point and confidence estimation. The reparametrization b 7 produced rather accurate confidence intervals. Exact confidence intervals for b in case of non-censoring were obtained using the pivotal quantity b/b.