Flexible Cure Rate Modeling Under Latent Activation Schemes

With rapid improvements in medical treatment and health care, many datasets dealing with time to relapse or death now reveal a substantial portion of patients who are cured (i.e., who never experience the event). Extended survival models called cure rate models account for the probability of a subject being cured and can be broadly classified into the classical mixture models of Berkson and Gage (BG type) or the stochastic tumor models pioneered by Yakovlev and extended to a hierarchical framework by Chen, Ibrahim, and Sinha (YCIS type). Recent developments in Bayesian hierarchical cure models have evoked significant interest regarding relationships and preferences between these two classes of models. Our present work proposes a unifying class of cure rate models that facilitates flexible hierarchical model-building while including both existing cure model classes as special cases. This unifying class enables robust modeling by accounting for uncertainty in underlying mechanisms leading to cure. Issues such as regressing on the cure fraction and propriety of the associated posterior distributions under different modeling assumptions are also discussed. Finally, we offer a simulation study and also illustrate with two datasets (on melanoma and breast cancer) that reveal our framework's ability to distinguish among underlying mechanisms that lead to relapse and cure.     

Maximum likelihood factor analysis of natural arthropod infestations in stored grain bulks

Summary Interrelations between natural arthropod fauna and their environment were studied in 1964 from 165 samples collected from the same number of boxcars each containing about 54 tons of stored grain originating from various parts of Western Canada. Applications of several multivariate procedures, such as principal component and maximum likelihood factor analyses (unrotated, rotated and oblique solutions) were applied to 6 biotic and 3 environmental variates to obtain the clearest picture of interrelations among these variates.J?reskog’s computational procedure and computer programs were used for unrotated, rotated and oblique maximum likelihood solutions. A 3-factor hypothesis was developed; of the 3 factors the first 2 were significantly correlated (Correlation coefficient=0.78±0.15). The final confirmatory factor analysis revealed that: 1) the first factor indicates that acarine variates,Glycyphagus andCheyletus, were correlated positively to each other but negatively with the insect Psocoptera, 2) the second factor indicates that the acarine variateTarsonemus is correlated to moisture, dockage and the entomological variate Psocoptera, 3) the third factor indicates that the environmental factors are correlated to each other, and 4) the correlations between the third factor and the other factors are statistically different from zero. Contribution no. 420 from the Research Station, Canada Department of Agriculture, Winnipeg, Manitoba.     

Youth and Religion: The Gameboy Generation Goes to “Church”

Using the secularization theory and the Marxistnotion of religion as masking class conscienceone would expect the importance of religion andreligious involvement today to wane and belimited to lower class members. To challenge thisexpectation, using a representative nationaltelephone survey of 2004 youth (ages 11–18) andtheir parents, we attempt to answer thefollowing two questions: How religious areteenagers, and what may explain variation inreligious perception and involvement amongteens. Findings suggest that religion remainsperceived as very important by most teenagersand parents report that about two-thirds ofteenagers attended a place of worship at leastmonthly and that two out of five attended asocial group sponsored by a religiousorganization. These findings do not support thesecularization theory. As expected, parentalattendance of religious worship, teen's age,and teen's ethnicity and gender weresignificantly associated with three variablesof religious behavior and attendance. Incontrast to the Marxist notion of religion,measures of socio-economic status indicatethat, in the contemporary United States,religious participation, but not beliefs, islargely the domain of the middle-upper classes.     

Analysis and simulation modelling of population dynamics and bioenergetics ofCryptolestes ferrugineus (Coleoptera: Cucujidae) in stored wheat

Summary The consequences of infestation of stored wheat by the rusty grain beetle,Cryptolestes ferrugineus (Stephens) was determined for 222 d at 30°C in 70-1 drums containing wheat at 13.5% moisture content. Temperature, grain moisture, seed damage, germination and weight, dust weight, fat acidity values (FAV), published data on growth, reproduction, survival and cannibalism rates and energy budget were used to develop a computer simulation model to simulate the population dynamics ofC. ferrugineus at 30°C. In the insect-free control system, the fungi,Alternaria alternata decreased,Aspergillus glaucus group andPenicillium spp. increased, probably causing a rise in FAV of the grain. In the insect-infested system,C. ferrugineus could only eat the wheat germ of kernels that had a broken bran layer; 35.7% of the wheat germ or 914.6 J per 100 kernels was consumed. Within two generations after initial introduction,C. ferrugineus reached a peak in numbers and biomass polluting the ecosystem with excreta and remains, and accelerating the deteriorative process observed in the insect-free control system by increasing respiration temperature, FAV and reducing grain germination. After 87 d, the insect population declined to low levels. The simulation model provided a close match between the observed and predicted numbers of insect life stages and bioenergetic variables during the insect population growth phase. Simulation trials suggested that cannibalism of larger compared with smaller immature stages would be more wasteful of developmental time and energy, reducing the number of individuals reaching reproductive age, and that density-dependent fecundity was probably not an important regulatory mechanism ofC. ferrugineus population dynamics in this study. Contribution No. 1314 from Agriculture Canada Research Station, Winnipeg, Manitoba, R3T 2M9, Canada     

Behrens-fisher problem under the assumption of homogeneous coefficients of variation

The powers of the likelihood ratio (LR) test and an “asymptotically (in some sense) optimum” invariant test are examined and compared by simulation techniques with those of several other relevant tests for the problem of testing the equality of two univariate normal population means under the assumption of heterogeneous variances but homogeneous coefficients of variation. It is seen that the LR test is highly satisfactory for all values of the coefficient of variation and the “asymptotically optimum” invariant test, which is computationally much simpler than the LR test, is a reasonably good competitor for cases where the value of the coefficient of variation is greater than or equal to 3. Also, a     

Do Parent Units Benefit from Reverse Knowledge Transfer?

Emerging market multinationals resort to knowledge acquisitions from their overseas subsidiaries to springboard and realize their global ambitions. Drawing from the knowledge‐based view and social capital perspective, this study explores the effects of organizational collaboration and tacitness on multiple dimensions of reverse knowledge transfer (RKT). Data were collected through a survey, from senior and middle level managers of parent Indian multinationals, pertaining to RKT from their overseas subsidiaries. The hypotheses are analysed using partial least squares modelling. The results demonstrate positive effects between the extent and benefits of RKT. Collaboration was found to have a positive influence on both dimensions of RKT. Tacitness also has a positive impact on the benefits from RKT. The implications of the findings and the limitations of the study are discussed along with suggestions for future research.     

Assessing Markups,Service Quality,and Product Attributes in Music CDs' Internet Retailing

We investigate trade‐offs among markups, service quality, and product attributes across customer, Internet retailer, and wholesaler echelons. Research has documented the reality of retail price dispersion, but little is known about how retail markups, in particular, are related to service quality and product attributes. For example, do Internet retailers deliver superior service in return for high markups? Do product characteristics affect the relationship between service and markups for retailers? To examine these issues, we first developed a model of Internet retail profitability that separates revenues and costs related to sales from other profit sources. This framework allowed us to position our work alongside the extant literature about Internet retailing. Moreover, it led us to synthesize service quality dimensions found in Internet retailing studies. We subsequently developed a critical‐event study based on the profit model and the synthesis of service quality dimensions to delineate service aspects that retailers should emphasize to address buyers' utility. Finally, we collected data from Internet purchases across retailers to isolate markup‐service quality trade‐offs along our delineated service aspects. We find that high markups are associated with superior performance across service quality dimensions. Furthermore, this trade‐off becomes more acutely defined when products with variable popularity are transacted.     

Nonnegative Unbiased Estimation of Scale Parameters and Associated Quantiles Based on a Ranked Set Sample

In this article, we are interested in estimating the scale parameter in location and scale families. It is well known that the best linear unbiased estimator (BLUE) of scale parameter based on a simple random sample (SRS) is nonnegative. However, the BLUE of scale parameter based on a ranked set sample (RSS) can assume negative values. We suggest various modifications of BLUE of scale parameter based on RSS so that the resulting estimators are unbiased as well as nonnegative. Their performances in terms of relative efficiencies are compared and some recommendations are made for normal, logistic, double exponential, two-parameter exponential and Weibull distributions. We also briefly discuss an application of the proposed nonnegative BLUE of scale parameter for quantile estimation for the above populations.     

Estimation of the common mean of two univariate normal populations

The problem of estimating the common mean μ of two univariate normal populations with unknown and unequal variances is considered from a decision-theoretic point of view. We restrict our attention to an appropriate class C and its three subclasses C_0C1C2of un-biased estimates of μ. We consider the usual estimate μ⁰ of μ which is the weighted linear combination of the sample means with weights as reciprocals of the sample variances. Its admissibility in C₀ and extended admissibility in C is proved. Admissible estimates in C₁ and C₂are also obtained.The loss is always assumed to be squared error. The question of admissibility of μ⁰ in the class of all estimators is still open.