Bayesian semiparametric zero-inflated Poisson model for longitudinal count data

This paper presents new methods, using a Bayesian approach, for analyzing longitudinal count data with excess zeros and nonlinear effects of continuously valued covariates. In longitudinal count data there are many problems that can make the use of a zero-inflated Poisson (ZIP) model ineffective. These problems are unobserved heterogeneity and nonlinear effects of continuously valued covariates. Our proposed semiparametric model can simultaneously handle these problems in a unified framework. The framework accounts for heterogeneity by incorporating random effects and has two components. The parametric component of the model which deals with the linear effects of time invariant covariates and the non-parametric component which gives an arbitrary smooth function to model the effect of time or time-varying covariates on the logarithm of mean count. The proposed methods are illustrated by analyzing longitudinal count data on the assessment of an efficacy of pesticides in controlling the reproduction of whitefly.     

The gamma-frailty Poisson model for the nonparametric estimation of panel count data

In this article, we study nonparametric estimation of the mean function of a counting process with panel observations. We introduce the gamma frailty variable to account for the intracorrelation between the panel counts of the counting process and construct a maximum pseudo-likelihood estimate with the frailty variable. Three simulated examples are given to show that this estimation procedure, while preserving the robustness and simplicity of the computation, improves the efficiency of the nonparametric maximum pseudo-likelihood estimate studied in Wellner and Zhang (2000, Annals of Statistics 28, 779-814). A real example from a bladder tumor study is used to illustrate the method.     

Generalized Poisson distribution: the property of mixture of Poisson and comparison with negative binomial distribution

We prove that the generalized Poisson distribution GP(theta, eta) (eta > or = 0) is a mixture of Poisson distributions; this is a new property for a distribution which is the topic of the book by Consul (1989). Because we find that the fits to count data of the generalized Poisson and negative binomial distributions are often similar, to understand their differences, we compare the probability mass functions and skewnesses of the generalized Poisson and negative binomial distributions with the first two moments fixed. They have slight differences in many situations, but their zero-inflated distributions, with masses at zero, means and variances fixed, can differ more. These probabilistic comparisons are helpful in selecting a better fitting distribution for modelling count data with long right tails. Through a real example of count data with large zero fraction, we illustrate how the generalized Poisson and negative binomial distributions as well as their zero-inflated distributions can be discriminated.     

Measuring the effect of including multiple cancers in survival analyses using data from the Canadian Cancer Registry

Background: In survival analyses using cancer registry data, second and subsequent primary cancers diagnosed in individuals are typically excluded. However, this approach may lead to biased comparisons of survival between cancer registries, or over time within a single registry. Purpose: To examine the impact of including multiple primary cancers in the derivation of survival estimates using data from a population-based national cancer registry. Methods: Five-year relative survival estimates for persons aged 15–99 years at diagnosis were derived using all eligible primary cases from the Canadian Cancer Registry (CCR)—a population-based registry containing information on cases diagnosed from 1992 onward—and then again using first primary cases only. Any pre-1992 cancer history of persons on the CCR was obtained by using auxiliary information. Results: The inclusion of multiple cancers resulted in lower estimates of 5-year relative survival for virtually all cancers studied. The effect was somewhat attenuated by age-standardization (e.g., from 1.3% to 1.0% for all cancers combined), and was greatest for bladder cancer (?2.4%) followed by oral cancer (?1.9%)—cancers that had the first and third lowest proportions of first cancers, respectively. For the majority of cancers the difference was less than 1.0%. Cancers for which there was virtually no difference (e.g., lung, pancreatic, ovarian and liver) tended to be those with a poor prognosis. Conclusion: Inclusion of second and subsequent primary cancers in the analysis tended to lower estimates of relative survival, the extent of which varied by cancer and age and depended in part on the proportion of first primary cancers.     

采用广义加法模型整合数字高程模型和遥感数据进行植被分布预测

为了采用广义加法模型整合数字高程模型和遥感数据进行植被分布的预测, 并探索耦合环境变量和遥感数据作为预测变量是否能够有效地提高植被分布预测的精度, 选择海拔、坡度、至黄河最近距离、至海岸线最近距离, 以及从SPOT5遥感影像中提取的光谱变量作为预测变量, 采用广义加法模型整合环境变量和光谱变量, 建立植被分布预测模型。研究设置3种建模情景(以环境变量作为预测变量, 以光谱变量作为预测变量, 综合使用环境变量与光谱变量作为预测变量)对黄河三角洲的优势植被类型的分布进行了预测, 并对预测结果采用偏差分析、受试者工作特征曲线和野外采样点对比等3种方法进行了验证。结果表明: (1)基于广义加法模型的植被分布预测方法具有一定的实用性, 可以较为准确地预测植被的分布; 盖度较高的植被类型预测精度较高, 盖度较低的植被类型预测精度较低, 植物群落结构的特点是出现这些差异的主要原因; 综合使用环境变量和光谱变量作为预测变量的模型, 预测精度高于单独以环境变量或者光谱变量作为预测变量的模型。(2)环境变量、光谱变量大多被选入模型, 二者均对植被分布预测有重要的作用; 同一预测变量在不同植被类型的预测模型中的贡献不同, 这与植被的光谱、环境特征差异有关; 同一预测变量在不同的建模情景下对模型的贡献不同, 环境变量与光谱变量的耦合效应可能是导致预测变量对模型的贡献出现变化的原因。     

The exponential‐Poisson model for recurrent event data: An application to a set of data on malaria in Brazil

In this paper, we introduce a new model for recurrent event data characterized by a baseline rate function fully parametric, which is based on the exponential‐Poisson distribution. The model arises from a latent competing risk scenario, in the sense that there is no information about which cause was responsible for the event occurrence. Then, the time of each recurrence is given by the minimum lifetime value among all latent causes. The new model has a particular case, which is the classical homogeneous Poisson process. The properties of the proposed model are discussed, including its hazard rate function, survival function, and ordinary moments. The inferential procedure is based on the maximum likelihood approach. We consider an important issue of model selection between the proposed model and its particular case by the likelihood ratio test and score test. Goodness of fit of the recurrent event models is assessed using Cox‐Snell residuals. A simulation study evaluates the performance of the estimation procedure in the presence of a small and moderate sample sizes. Applications on two real data sets are provided to illustrate the proposed methodology. One of them, first analyzed by our team of researchers, considers the data concerning the recurrence of malaria, which is an infectious disease caused by a protozoan parasite that infects red blood cells.     

Mast cell tumours and other skin neoplasia in Danish dogs - data from the Danish Veterinary Cancer Registry

Background  

The Danish Veterinary Cancer Registry (DVCR) was established in May 2005 to gather information about neoplasms in the Danish dog and cat populations. Practitioners from more than 60 clinics throughout Denmark have submitted data on these species. The objectives of the current study were, with a special focus on mast cell tumours (MCT) to investigate the occurrence, gender distribution, biological behaviour, locations, types, the diagnostic method used and treatment of skin neoplasms in dogs based on information reported to the DVCR.     

Combining Information from Cancer Registry and Medical Records Data to Improve Analyses of Adjuvant Cancer Therapies

Summary .  Cancer registry records contain valuable data on provision of adjuvant therapies for cancer patients. Previous studies, however, have shown that these therapies are underreported in registry systems. Hence direct use of the registry data may lead to invalid analysis results. We propose first to impute correct treatment status, borrowing information from an additional source such as medical records data collected in a validation sample, and then to analyze the multiply imputed data, as in Yucel and Zaslavsky (2005,  Journal of the American Statistical Association   100, 1123–1132). We extend their models to multiple therapies using multivariate probit models with random effects. Our model takes into account the associations among different therapies in both administration and probability of reporting, as well as the multilevel structure (patients clustered within hospitals) of registry data. We use Gibbs sampling to estimate model parameters and impute treatment status. The proposed methodology is applied to the data from the Quality of Cancer Care project, in which stage II or III colorectal cancer patients were eligible to receive adjuvant chemotherapy and radiation therapy.     

Applicability of broken-rodlike chain model to conformational analysis of polypeptide chain

In order to check the applicability of the broken-rodlike (BR) chain model, consisting of several rods alternatively joined by flexible random coils, to the conformational analysis of a polypeptide chain in the helix-to-coil transition regions, two relations predicted by the Zimm and Bragg theory and the method with the BR chain model are compared. It is shown that, despite a clear difference between the models employed in the two methods, they give substantially identical results in both probability P(j) that a helical residue is in a helical sequence j units long and averaged helical fraction dependence of the mean-squared radius of gyration. Thus the use of the method with the BR chain model in the conformational analysis of a polypeptide chain could be rationalized, at least, with the same degree of approximation as is assumed in the Zimm and Bragg theory. Using the scattering function for the BR chain model, averaged helical-sequence lengths are evaluated for partially ionized poly(L-glutamic acid) (PGA) in added-salt aqueous solution and nonionized PGA in N-methylacetamide, both in a helical state. As a result, it is shown that the length in the latter molecule is approximately tenfold longer than that in the former one.     

Background stratified Poisson regression analysis of cohort data

Background stratified Poisson regression is an approach that has been used in the analysis of data derived from a variety of epidemiologically important studies of radiation-exposed populations, including uranium miners, nuclear industry workers, and atomic bomb survivors. We describe a novel approach to fit Poisson regression models that adjust for a set of covariates through background stratification while directly estimating the radiation-disease association of primary interest. The approach makes use of an expression for the Poisson likelihood that treats the coefficients for stratum-specific indicator variables as ‘nuisance’ variables and avoids the need to explicitly estimate the coefficients for these stratum-specific parameters. Log-linear models, as well as other general relative rate models, are accommodated. This approach is illustrated using data from the Life Span Study of Japanese atomic bomb survivors and data from a study of underground uranium miners. The point estimate and confidence interval obtained from this ‘conditional’ regression approach are identical to the values obtained using unconditional Poisson regression with model terms for each background stratum. Moreover, it is shown that the proposed approach allows estimation of background stratified Poisson regression models of non-standard form, such as models that parameterize latency effects, as well as regression models in which the number of strata is large, thereby overcoming the limitations of previously available statistical software for fitting background stratified Poisson regression models.     

Some goodness-of-fit methods for the Poisson plus added zeros distribution

Methods for making inferences about the Poisson plus added zeros distribution and the truncated Poisson distribution are presented and illustrated with bacteriological data. Some of the methods are designed for testing the compatibility of the zero frequency with the Poisson distribution, whereas others are given for testing the goodness of fit for the truncated Poisson. In particular, a modified form of the Fisher index of dispersion is presented which is suitable for the truncated case. It is shown that the use of the usual expression of the index of dispersion for testing the adequacy of the truncated Poisson is not correct and leads to accepting inadequate fits more frequently than expected on the basis of test of significance. Furthermore, three test statistics are presented for testing the compatability of the zero frequency with the Poisson distribution. The results of the simulation show that two test statistics, one due to Cochran (W. G. Cochran, Biometrics 10:417-451, 1954) and the other to Rao and Chakravarti (C. R. Rao and I. M. Chakravarti, Biometrics 12:264-282, 1956), are preferable to those from the likelihood ratio test.     

基于地理加权泊松模型的天然次生林进界株数空间分布与预测

基于帽儿山实验林场2004—2016年森林资源二类调查固定样地(共108块)数据,采用全局Poisson模型和4种空间尺度(2.5、5、10、15 km)下的地理加权泊松模型(geographically weighted Poisson regression, GWPR)对天然次生林进界株数的空间分布进行了研究,并对5种模型的拟合效果以及影响林分进界株数的因子进行了分析,利用莫兰指数描述了模型残差在全局和局域两种水平上的空间自相关性.结果表明: 本文所选的林分及地形因子都显著影响天然次生林进界株数的空间分布,林分平均胸径是最主要的影响因子;在小尺度(2.5 km)下GWPR模型拥有很高的拟合精度,产生了最大范围的模型参数估计值,得到了较好的模型参数局域化空间分布效果;在较小尺度(2.5和5 km)下GWPR模型产生了较小范围的模型残差,模型的稳定性得到提升;在小尺度(2.5 km)下GWPR模型残差的全局空间自相关性达到最低,局域空间自相关性显著减小,并形成了不同观测值少量聚类这一理想的空间分布模式;在对进界株数空间分布的模拟效果上,小尺度(2.5 km)下的局域模型明显好于全局模型.     

Cancer incidence in Iran in 2014: Results of the Iranian National Population-based Cancer Registry

BackgroundWe aimed to report, for the first time, the results of the Iranian National Population-based Cancer Registry (INPCR) for the year 2014.MethodsTotal population of Iran in 2014 was 76,639,000. The INPCR covered 30 out of 31 provinces (98% of total population). It registered only cases diagnosed with malignant new primary tumors. The main sources for data collection included pathology center, hospitals as well as death registries. Quality assessment and analysis of data were performed by CanReg-5 software. Age standardized incidence rates (ASR) (per 100,000) were reported at national and subnational levels.ResultsOverall, 112,131 new cancer cases were registered in INPCR in 2014, of which 60,469 (53.9%) were male. The diagnosis of cancer was made by microscopic confirmation in 76,568 cases (68.28%). The ASRs of all cancers were 177.44 and 141.18 in male and female, respectively. Cancers of the stomach (ASR = 21.24), prostate (18.41) and colorectum (16.57) were the most common cancers in men and the top three cancers in women were malignancies of breast (34.53), colorectum (11.86) and stomach (9.44). The ASR of cervix uteri cancer in women was 1.78. Our findings suggested high incidence of cancers of the esophagus, stomach and lung in North/ North West of Iran.ConclusionOur results showed that Iran is a medium-risk area for incidence of cancers. We found differences in the most common cancers in Iran comparing to those reported for the World. Our results also suggested geographical diversities in incidence rates of cancers in different subdivisions of Iran.