Markov structure based logistic regression for repeated measures: An application to diabetes mellitus data

In longitudinal studies, as repeated observations are made on the same individual the response variables will usually be correlated. In analyzing such data, this dependence must be taken into account to avoid misleading inferences. The focus of this paper is to apply a logistic marginal model with Markovian dependence proposed by Azzalini [A. Azzalini, Logistic regression for autocorrelated data with application to repeated measures, Biometrika 81 (1994) 767–775] to the study of the influence of time-dependent covariates on the marginal distribution of the binary response in serially correlated binary data. We have shown how to construct the model so that the covariates relate only to the mean value of the process, independent of the association parameters. After formulating the proposed model for repeated measures data, the same approach is applied to missing data. An application is provided to the diabetes mellitus data of registered patients at the Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM) in 1984, using both time stationary and time varying covariates.     

Prediction of risks of sequence of events using multistage proportional hazards model: a marginal-conditional modelling approach

Statistical Methods & Applications - In many studies, sequence of events may occur over time that produce repeated measures with censored observations. Multi-state models are commonly used, and...     

Multiple linear regression model with stochastic design variables

In a simple multiple linear regression model, the design variables have traditionally been assumed to be non-stochastic. In numerous real-life situations, however, they are stochastic and non-normal. Estimators of parameters applicable to such situations are developed. It is shown that these estimators are efficient and robust. A real-life example is given.     

Nonparametric versus Parametric Estimation of the Cure Fraction Using Interval Censored Data

This article discusses estimation of the cure rate by means of the bounded cumulative hazard (BCH) model using interval censored data. The parametric and nonparametric estimation methods within the framework of the EM algorithm were employed for cure rate estimation and their results compared. The Turnbull estimator was used in the nonparametric estimation while in parametric method both the exponential and Weibull distributions were considered. We show via simulation that the nonparametric method is a viable alternative to the parametric one when the censoring rate is rapidly increasing.     

Soil health as a predictor of lettuce productivity and quality: A case study of urban vacant lots

Urban agriculture offers a framework for local self-reliance and resilience in cities. However, there is a concern over the capacity of urban soil to provide sustainable and safe food production. We tested the effectiveness of several soil health indicators to predict food crop productivity and quality in vacant lots in a disadvantaged neighborhood in the city of Cleveland, Ohio. We defined soil health as a state of composite well being in terms of biological, chemical, and physical properties of the soil as they relate to crop productivity. Twelve city-owned vacant lots, three close to each of the four city schools, were selected for soil properties and plant growth analyses. Soil samples were analyzed for pH, moisture content (θv), soil texture, soil organic matter (SOM), active carbon (AC), ammonium (NH₄-N), nitrate (NO₃-N), microbial biomass N (MBN), and nematode community parameters including total (TNN), bacteria-feeding (BFN), fungal-feeding (FFN), and plant-parasitic (PPN) nematodes, number of nematode genera (NNG), and nematode food web enrichment index (EI) and structure index (SI). Lettuce was planted in the selected vacant lots and its growth was documented through measures of dry biomass, numbers of leaves/plant, and complementary subjective appearance scores related to physiological status. All measured parameters varied considerably among vacant lots except soil pH. Principal components analysis revealed that among the primary soil physical, chemical, and biological parameters, soil clay, NO₃-N, MBN, SOM, AC, TNN, BFN, FFN, and PPN contributed most to the variance of the entire dataset. There were also several positive correlations among these key soil health predictor variables: AC was positively correlated with clay, SOM, MBN, TNN, BFN, FFN and PPN, and TNN was positively correlated with AC, SOM, MBN, BFN, FFN and PPN. Of the identified primary soil health indicators, only clay, SOM, and MBN positively correlated with lettuce dry biomass, which was also positively correlated with a secondary soil health indicator, the nematode food web EI. Lettuce leaf necrosis was negatively correlated with clay, AC, SOM, MBN, TNN, FFN, and PPN, and the proportion of withered leaves was negatively correlated only with SOM. It is concluded that AC, PPN, TNN, SOM, MBN, clay, and nematode food web EI can serve as important soil health indicators that have potential for predicting crop productivity and quality in urban soils. It is also concluded that lettuce can serve as an important indicator of soil health with respect to crop productivity and quality in vacant lots.     

Joint estimation of the mean and dispersion parameters in the analysis of proportions: A comparison of efficiency and bias

This paper deals with joint estimation of the mean and dispersion parameters in the analysis of proportions. We consider a parametric model, namely the extended beta-binomial model, and several semiparametric procedures. We study large-sample efficiency and small-sample bias and efficiency properties of the estimates of the mean and intraclass correlation parameters. Estimation and efficiency calculations are présentés for the regression model. However, for simplicity, numerical large-sample efficiency and small-sample bias and efficiency calculations are performed for the two-parameter model only. Numerical efficiency results are présentés in terms of graphs. Estimated asymptotic efficiencies of various estimates are also compared for two data sets. Our findings suggest that for the estimation of the mean (regression) parameters the quasilikelihood procedure performs best. However, for the joint estimation, the Gaussian likelihood estimates perform best.     

Uptake of Chloroform by Skin During Short Exposures to Contaminated Water

Uptake of chloroform into hairless rat stratum corneum from dilute aqueous solutions was studied using tape-stripping to determine amounts deposited in the skin under various environmental exposure scenarios. The length of exposure of sedated animals to the chloroform-containing medium, the frequency and duration of tape-stripping, and the number of tape-strips per location were varied to map the stratum corneum substantivity of chloroform. Eight minutes immersion of the rat within a well-stirred solution at 36°C was found to be adequate time for the gradient to be established fully across the stratum corneum. Penetration was progressively deeper as the exposure time increased. Substantial evaporative loss of chloroform from the aqueous medium of application seem to be responsible for lower cumulative amounts taken up when the same solution was held on the rat's skin within a stainless steel template of fixed area. Of the total uptake (29 mg) from a dilute stirred solution of chloroform (0.44 mg/ml) at 36°C, about 95% was systematically absorbed after a 30 min exposure as determined by residuals (measurement of bath concentrations).     

Regional disparities in Indonesia: A social indicators approach

The paper examines the nature of regional imbalance in Indonesia by using a number of social indicators. Several disaggregated social indices are constructed with the help of a taxonomic technique. For the purpose of comparison, an overall economic index is also constructed. It is observed that the ranking of provinces differ substantially across the indices, implying that there is no causal relationship between the selected social and economic indicators. Regional disparities declined in the past decade in terms of all social indices except the demographic index. Economic inequality nowever increased between 1970 and 1980.