Right-Censored Time Series Modeling by Modified Semi-Parametric A-Spline Estimator

This paper focuses on the adaptive spline (A-spline) fitting of the semiparametric regression model to time series data with right-censored observations. Typically, there are two main problems that need to be solved in such a case: dealing with censored data and obtaining a proper A-spline estimator for the components of the semiparametric model. The first problem is traditionally solved by the synthetic data approach based on the Kaplan–Meier estimator. In practice, although the synthetic data technique is one of the most widely used solutions for right-censored observations, the transformed data’s structure is distorted, especially for heavily censored datasets, due to the nature of the approach. In this paper, we introduced a modified semiparametric estimator based on the A-spline approach to overcome data irregularity with minimum information loss and to resolve the second problem described above. In addition, the semiparametric B-spline estimator was used as a benchmark method to gauge the success of the A-spline estimator. To this end, a detailed Monte Carlo simulation study and a real data sample were carried out to evaluate the performance of the proposed estimator and to make a practical comparison.

Zero-inflated-censored Weibull and gamma regression models to estimate wild boar population dispersal distance

The dynamics of the wild boar population has become a pressing issue not only for ecological purposes, but also for agricultural and livestock production. The data related to the wild boar dispersal distance can have a complex structure, including excess of zeros and right-censored observations, thus being challenging for modeling. In this sense, we propose two different zero-inflated-right-censored regression models, assuming Weibull and gamma distributions. First, we present the construction of the likelihood function, and then, we apply both models to simulated datasets, demonstrating that both regression models behave well. The simulation results point to the consistency and asymptotic unbiasedness of the developed methods. Afterwards, we adjusted both models to a simulated dataset of wild boar dispersal, including excess of zeros, right-censored observations, and two covariates: age and sex. We showed that the models were useful to extract inferences about the wild boar dispersal, correctly describing the data mimicking a situation where males disperse more than females, and age has a positive effect on the dispersal of the wild boars. These results are useful to overcome some limitations regarding inferences in zero-inflated-right-censored datasets, especially concerning the wild boar’s population. Users will be provided with an R function to run the proposed models.

Hydrochemical characteristics and irrigation quality of groundwater in the Sosnovskiy district of the Chelyabinsk region

The purpose of the study was to assess the quality of groundwater used for irrigation of individual plots of land, belonging to the residents of the Sosnovskiy district of the Chelyabinsk region. Material: 73 samples taken in wells and boreholes on the territory of private land plots, villages, villa communities and garden non-profit partnerships. Methods: 30 indicators of water composition were determined (common indicators – 6, major ions and carbonate ion – 8, compounds of nitrogen, phosphorus and organic matter – 5, heavy metals and silicon – 11) using capillary electrophoresis, atomic absorption, potentiometry, spectrophotometry and conductometry. Statistical analysis included: estimation of means with 95 % confidence intervals (95 % CI) calculated by bootstrap (Kaplan – Meier method was used for censored observations), relative frequencies (in percent) with 95 % CI (Jeffreys method and simultaneous CI by Wilson). Data was visualized using Piper’s and Gibbs’ diagrams. Software: PAST, Scout 2008, R, DescTool, GW_Chart, KyPlot, QGIS. Results. On the territory of the Sosnovsky district of the Chelyabinsk region, (Ca2+–Mg2+ and HCO3––CO32–) type groundwater prevails with a pH 7.39 (95 % CI: from 7.30 to 7.48) and a TDS 651 mg/dm3 (95 % CI: from 598 to 720). Using Piper and Gibbs diagrams, it was established that the leading processes of the water chemical composition formation are: 1) the dissolution of the calcite-dolomite series’ minerals of sedimentary rocks, 2) the dissolution and the ion exchange of feldspars in the composition of granites with calcite. For mineralization and risks of chloride salinization, sodium salinization, magnesium salinization and soda formation, the proportion of samples with quality for irrigation purposes I and II category is 65.7 %. There are excesses of standards in the concentration of NO3- (24.7 % of samples), Ba (68.3 %), Cd (30.4 %), Fe (15.3 %), Mn (12.9 %), Pb (6.3 %).

A New Method for Imputing Censored Values in Crossover Designs with Time-to-Event Outcomes Using Median Residual Life

Crossover designs are commonly applied in research due to efficiency and subject parsimony compared to parallel studies. Baseline measurements would improve the power of comparison. For time to event outcomes, the sample size is reduced due to censorship, if they are ignored; thus, applying traditional regression models will be limited. A logical solution is to impute the censored observation and apply common analytical models for analyzing the data. Nevertheless, techniques to impute censored data in time-to-event outcomes in crossover designs are not practiced as much. Accordingly, we propose a method to impute the censored observation using median residual life regression and then analyze the data using analyses of covariance (ANCOVA), considering the difference of period-specific baselines as covariate. We used simulation to show the favorable performance of our method relative to a recently proposed method, multiple imputation with model averaging and ANCOVA (MIMI). Specifically, the censored observations were multiply-imputed using prespecified parametric event time models, and then, the methods were applied to a real data example.

Statistical deconvolution for inference of infection time series

Accurate measurement of daily infection incidence is crucial to epidemic response. However, delays in symptom onset, testing, and reporting obscure the dynamics of transmission, necessitating methods to remove the effects of stochastic delays from observed data. Existing estimators can be sensitive to model misspecification and censored observations; many analysts have instead used methods that exhibit strong bias or do not account for delays. We develop an estimator with a regularization scheme to cope with these sources of noise, which we term the Robust Incidence Deconvolution Estimator (RIDE). We validate RIDE on synthetic data, comparing accuracy and stability to existing approaches. We then use RIDE to study COVID-19 records in the United States, and find evidence that infection estimates from reported cases can be more informative than estimates from mortality data. To implement these methods, we release incidental, a ready-to-use R implementation of our estimator that can aid ongoing efforts to monitor the COVID-19 pandemic.

Left-censored dementia incidences in estimating cohort effects

We estimate the dementia incidence hazard in Germany for the birth cohorts 1900 until 1954 from a simple sample of Germany’s largest health insurance company. Followed from 2004 to 2012, 36,000 uncensored dementia incidences are observed and further 200,000 right-censored insurants included. From a multiplicative hazard model we find a positive and linear trend in the dementia hazard over the cohorts. The main focus of the study is on 11,000 left-censored persons who have already suffered from the disease in 2004. After including the left-censored observations, the slope of the trend declines markedly due to Simpson’s paradox, left-censored persons are imbalanced between the cohorts. When including left-censoring, the dementia hazard increases differently for different ages, we consider omitted covariates to be the reason. For the standard errors from large sample theory, left-censoring requires an adjustment to the conditional information matrix equality.