The Kumaraswamy Weibull distribution with application to failure data

For the first time, we introduce and study some mathematical properties of the Kumaraswamy Weibull distribution that is a quite flexible model in analyzing positive data. It contains as special sub-models the exponentiated Weibull, exponentiated Rayleigh, exponentiated exponential, Weibull and also the new Kumaraswamy exponential distribution. We provide explicit expressions for the moments and moment generating function. We examine the asymptotic distributions of the extreme values. Explicit expressions are derived for the mean deviations, Bonferroni and Lorenz curves, reliability and Renyi entropy. The moments of the order statistics are calculated. We also discuss the estimation of the parameters by maximum likelihood. We obtain the expected information matrix. We provide applications involving two real data sets on failure times. Finally, some multivariate generalizations of the Kumaraswamy Weibull distribution are discussed. (C) 2010 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.

A log-extended Weibull regression model

A bathtub-shaped failure rate function is very useful in survival analysis and reliability studies. The well-known lifetime distributions do not have this property. For the first time, we propose a location-scale regression model based on the logarithm of an extended Weibull distribution which has the ability to deal with bathtub-shaped failure rate functions. We use the method of maximum likelihood to estimate the model parameters and some inferential procedures are presented. We reanalyze a real data set under the new model and the log-modified Weibull regression model. We perform a model check based on martingale-type residuals and generated envelopes and the statistics AIC and BIC to select appropriate models. (C) 2009 Elsevier B.V. All rights reserved.

The generalized log-gamma mixture model with covariates: local influence and residual analysis

In a sample of censored survival times, the presence of an immune proportion of individuals who are not subject to death, failure or relapse, may be indicated by a relatively high number of individuals with large censored survival times. In this paper the generalized log-gamma model is modified for the possibility that long-term survivors may be present in the data. The model attempts to separately estimate the effects of covariates on the surviving fraction, that is, the proportion of the population for which the event never occurs. The logistic function is used for the regression model of the surviving fraction. Inference for the model parameters is considered via maximum likelihood. Some influence methods, such as the local influence and total local influence of an individual are derived, analyzed and discussed. Finally, a data set from the medical area is analyzed under the log-gamma generalized mixture model. A residual analysis is performed in order to select an appropriate model.

Bias-corrected Pearson estimating functions for Taylor`s power law applied to benthic macrofauna data

Estimation of Taylor`s power law for species abundance data may be performed by linear regression of the log empirical variances on the log means, but this method suffers from a problem of bias for sparse data. We show that the bias may be reduced by using a bias-corrected Pearson estimating function. Furthermore, we investigate a more general regression model allowing for site-specific covariates. This method may be efficiently implemented using a Newton scoring algorithm, with standard errors calculated from the inverse Godambe information matrix. The method is applied to a set of biomass data for benthic macrofauna from two Danish estuaries. (C) 2011 Elsevier B.V. All rights reserved.

Influence diagnostics for polyhazard models in the presence of covariates

In this paper, we present various diagnostic methods for polyhazard models. Polyhazard models are a flexible family for fitting lifetime data. Their main advantage over the single hazard models, such as the Weibull and the log-logistic models, is to include a large amount of nonmonotone hazard shapes, as bathtub and multimodal curves. Some influence methods, such as the local influence and total local influence of an individual are derived, analyzed and discussed. A discussion of the computation of the likelihood displacement as well as the normal curvature in the local influence method are presented. Finally, an example with real data is given for illustration.

Log-modified Weibull regression models with censored data: Sensitivity and residual analysis

This paper proposes a regression model considering the modified Weibull distribution. This distribution can be used to model bathtub-shaped failure rate functions. Assuming censored data, we consider maximum likelihood and Jackknife estimators for the parameters of the model. We derive the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes and we also present some ways to perform global influence. Besides, for different parameter settings, sample sizes and censoring percentages, various simulations are performed and the empirical distribution of the modified deviance residual is displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extended for a martingale-type residual in log-modified Weibull regression models with censored data. Finally, we analyze a real data set under log-modified Weibull regression models. A diagnostic analysis and a model checking based on the modified deviance residual are performed to select appropriate models. (c) 2008 Elsevier B.V. All rights reserved.

On estimation and influence diagnostics for zero-inflated negative binomial regression models

The zero-inflated negative binomial model is used to account for overdispersion detected in data that are initially analyzed under the zero-Inflated Poisson model A frequentist analysis a jackknife estimator and a non-parametric bootstrap for parameter estimation of zero-inflated negative binomial regression models are considered In addition an EM-type algorithm is developed for performing maximum likelihood estimation Then the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes and some ways to perform global influence analysis are derived In order to study departures from the error assumption as well as the presence of outliers residual analysis based on the standardized Pearson residuals is discussed The relevance of the approach is illustrated with a real data set where It is shown that zero-inflated negative binomial regression models seems to fit the data better than the Poisson counterpart (C) 2010 Elsevier B V All rights reserved

Double generalized linear model for tissue culture proportion data: a Bayesian perspective

Joint generalized linear models and double generalized linear models (DGLMs) were designed to model outcomes for which the variability can be explained using factors and/or covariates. When such factors operate, the usual normal regression models, which inherently exhibit constant variance, will under-represent variation in the data and hence may lead to erroneous inferences. For count and proportion data, such noise factors can generate a so-called overdispersion effect, and the use of binomial and Poisson models underestimates the variability and, consequently, incorrectly indicate significant effects. In this manuscript, we propose a DGLM from a Bayesian perspective, focusing on the case of proportion data, where the overdispersion can be modeled using a random effect that depends on some noise factors. The posterior joint density function was sampled using Monte Carlo Markov Chain algorithms, allowing inferences over the model parameters. An application to a data set on apple tissue culture is presented, for which it is shown that the Bayesian approach is quite feasible, even when limited prior information is available, thereby generating valuable insight for the researcher about its experimental results.

Distribution of Aceria guerreronis and Neoseiulus baraki among and within coconut bunches in northeast Brazil

Aceria guerreronis Keifer (Acari: Eriophyidae) is considered a major pest of coconut in many countries in the Americas, Africa and parts of Asia. Neoseiulus baraki Athias-Henriot (Acari: Phytoseiidae) is one of the predatory mites most commonly found in association with A. guerreronis in parts of northeast Brazil. The objective of this work was to study the distribution of A. guerreronis and N. baraki among and within coconut bunches. The hypothesis was tested that A. guerreronis and N. baraki are homogenously distributed over the fruits in a bunch, independent of the fruits` age and position. Five collections of bunches, each corresponding to leaves 12-16 from apex (about 2-6 month-old), were conducted in each of three fields in northeastern Brazil, from February to October, 2007. A total of 1,986 fruits were examined. The number of mites, the percentage of fruits hosting them and the level of damage caused by A. guerreronis were evaluated. The highest density of A. guerreronis was observed on fruits of bunch 4 whereas the highest density of N. baraki was observed on bunch 5. Considering all fruits together, no significant differences were observed between densities of either A. guerreronis or N. baraki among the basal, median and apical thirds of the bunches. In younger bunches, fruits of the apical region tend to have lower densities of both mites than fruits of the basal region. This pattern, in association with a similar pattern for the percentage of fruits hosting N. baraki, suggests that the predator initially reaches the basal bunch region, from where it moves to the apical region. The results of the present study suggest that the pest population reduction in bunches older than bunch 4 could be due to (1) an effect of the predator, (2) reduction of the proportion of undamaged tissues amenable to attack, and/or (3) less favorable characteristics of the fruits to attack by A. guerreronis, as indicated by their increasing lignin content as they get older.

Modelling the potential distribution of the invasive tomato red spider mite, Tetranychus evansi (Acari: Tetranychidae)

Predicting the potential geographical distribution of a species is particularly important for pests with strong invasive abilities. Tetranychus evansi Baker & Pritchard, possibly native to South America, is a spider mite pest of solanaceous crops. This mite is considered an invasive species in Africa and Europe. A CLIMEX model was developed to predict its global distribution. The model results fitted the known records of T. evansi except for some records in dry locations. Dryness as well as excess moisture stresses play important roles in limiting the spread of the mite in the tropics. In North America and Eurasia its potential distribution appears to be essentially limited by cold stress. Detailed potential distribution maps are provided for T. evansi in the Mediterranean Basin and in Japan. These two regions correspond to climatic borders for the species. Mite establishment in these areas can be explained by their relatively mild winters. The Mediterranean region is also the main area where tomato is grown in open fields in Europe and where the pest represents a threat. According to the model, the whole Mediterranean region has the potential to be extensively colonized by the mite. Wide expansion of the mite to new areas in Africa is also predicted. Agricultural issues highlighted by the modelled distribution of the pest are discussed.

Distribution of Tetranychus evansi and its predator Phytoseiulus longipes (Acari : Tetranychidae, Phytoseiidae) in southern Brazil

Tetranychus evansi is an important pest of tomato in several countries. The predatory mite Phytoseiulus longipes has been found in association with it in Uruguaiana, State of Rio Grande do Sul, Brazil. The objective of this study was to evaluate the distribution of those two species in Uruguaiana region. Bi-weekly samples of plant parts were taken between January 28 and April 30, 2007 in Uruguaiana and five neighboring counties. Tetranychus evansi was found in all counties, but P. longipes was only found in Uruguaiana, mainly in the urban area. It is conceivable that such restricted distribution of P. longipes is due to its recent introduction to the region. Alternatively, unfavorable winter conditions could prevent its persistence, except in protected places. Buildings and vegetation characteristics of the urban area of Uruguaiana could moderate strong winds, intense rainfall and low temperatures that commonly occur in the area.

Wing polymorphism and dispersal of Scaptocoris carvalhoi (Hemiptera : Cydnidae)

Functional wing polymorphism is commonly observed it) insects, and it may confer an important adaptive value to populations that bear this trait, because it allows dispersal and the location to more favorable habitats for their survival and reproduction. According to the oogenesis-flight syndrome theory, such wing polymorphism may imply differences in the locomotion Capacity of individuals, which is a factor induced by adverse environmental conditions during muscle development in immatures. Scaptocoris carvalhoi Becker (Hemiptera: Cydnidae) is an important agriculture pest in Brazil, and it has burrowing habits. The adults swarm in the beginning of the rainy season after a prolonged drought period in the Brazilian cerrado region. In these swarms, part of the population leaves the soil, performing long flights until locations with more abundant vegetation. In this study, we characterized wing polymorphism in S. carvalhoi, this being the first description in a species of Cydnidae. Brachypterous and macropterous males and females were observed, which showed positive and significant correlations between body length and hindwing length. Macropterous individuals demonstrated greater locomotion capacity than brachypterous individuals. In addition, only long-winged adults could fly, showing wing mobility and flight reaction. The increased number of macropterous individuals inside the soil during the swarming season and in the beginning of the rainy period suggests that wing polymorphism in this population occurs in seasonal cycles and that factors related to the scarcity of rains influence the development of immatures and the formation of polymorphic adults.

Biogeographic distribution of the phytoseiidae (Acari : Mesostigmata)

More than 1982 species in 90 genera were included in an analysis of the biogeography of the Phytoseiidae, a family of predatory mites. Seven biogeographic regions were taken into account: Nearctic, Neotropical, Ethiopian, West Palaearctic, East Palaearctic, Oriental, and Australasian. The number of species was particularly high in the Neotropical, Oriental, and West Palaearctic regions. These regions also present the highest levels of species endemism. The number of genera was quite similar in all regions except for the Neotropics, which also had a high level of endemism. The possible Gondwanian (Neotropical, Ethiopian, Australasian, and Oriental regions) origin of the Phytoseiidae, most probably in the Neotropics, and their possible radiation to Laurasia (Nearctic, West Palaearctic, and East Palaearctic regions) are discussed. The comparison between genera and species in the different biogeographic regions indicate the importance of both dispersal and vicariance events in the evolution of the group. Dispersal is assumed to have been most important between Neotropical and Nearctic regions and between East Palaearctic and Oriental regions, whereas vicariance could have been the dominating process between Australasian, Ethiopian, and Oriental regions, as well as between West and East Palaearctic regions. A parsimony analysis of endemicity showed the Neotropical and the Nearctic regions to be isolated from the other regions. This is certainly due to a diversification after the continents drifted apart and then a high dispersal between Nearctic and Neotropical regions. Different phylogenetic hypotheses and scenarios are proposed for each subfamily based on the results obtained and further investigations are proposed. (C) 2008 The Linnean Society of London.

Distribution and storage characterization of soil solution for drip irrigation

The increased use of marginal quality water with drip irrigation requires sound fertigation practices that reconcile environmental concerns with viable crop production objectives. We conducted experiments to characterize dynamics and patterns of soil solution within wet bulb formed by drip irrigation. Time-domain reflectometry probes were used to monitor the distribution of potassium nitrate (KNO(3)) and water distribution from drippers discharging at constant flow rates of 2, 4 and 8 L h(-1) in soil-filled containers. Considering results from different profiles, we observed greater solute storage near the dripper decreasing gradually towards the wetting front. About half of the applied KNO(3) solution (48%) was stored in the first layer (0-0.10 m) for all experiments, 29% was stored in the next layer (0.10-0.20 m). Comparing different dripper flow rates, we observed higher solution storage for 4 L h(-1), with 45, 53 and 47% of applied KNO(3) solution accumulating in the first layer (0-0.10 m) for dripper flow rates of 2, 4 and 8 L h(-1), respectively. The results suggest that based on the volume and frequency used in this experiment, it would be advantageous to apply small amounts of solution at more frequent intervals to reduce deep percolation losses of applied water and solutes.

Macroscopic root water uptake distribution using a matric flux potential approach

Hydrological models featuring root water uptake usually do not include compensation mechanisms such that reductions in uptake from dry layers are compensated by an increase in uptake from wetter layers. We developed a physically based root water uptake model with an implicit compensation mechanism. Based on an expression for the matric flux potential (M) as a function of the distance to the root, and assuming a depth-independent value of M at the root surface, uptake per layer is shown to be a function of layer bulk M, root surface M, and a weighting factor that depends on root length density and root radius. Actual transpiration can be calculated from the sum of layer uptake rates. The proposed reduction function (PRF) was built into the SWAP model, and predictions were compared to those made with the Feddes reduction function (FRF). Simulation results were tested against data from Canada (continuous spring wheat [(Triticum aestivum L.]) and Germany (spring wheat, winter barley [Hordeum vulgare L.], sugarbeet [Beta vulgaris L.], winter wheat rotation). For the Canadian data, the root mean square error of prediction (RMSEP) for water content in the upper soil layers was very similar for FRF and PRF; for the deeper layers, RMSEP was smaller for PRF. For the German data, RMSEP was lower for PRF in the upper layers and was similar for both models in the deeper layers. In conclusion, but dependent on the properties of the data sets available for testing,the incorporation of the new reduction function into SWAP was successful, providing new capabilities for simulating compensated root water uptake without increasing the number of input parameters or degrading model performance.