Environmetrics最新文献

We propose a Poisson model for zero-inflated spatial counts contaminated by measurement error: we accommodate the excess of zeroes in the counts, consider the possible under/over reporting of the response and account for the neighboring structure of spatial areal units. Bayesian inferences are provided by MCMC implementation through the R package NIMBLE. To evaluate the model performance, a simulation study is carried out under configurations that allow for structured and unstructured spatial random effects. The proposed model is applied to investigate the distribution of the counts of wildfire occurrences in the municipal areas of two neighboring Italian regions for the summer season 2021. Fire counts are obtained by processing MODIS satellite data, while several socio-economic and environmental-driven potential risk factors are also considered in the model formulation. Data from multiple sources with different spatial support are processed in order to comply with the municipal units. Results suggest the appropriateness of the approach and provide some insights on the features of wildfire occurrences.

Data depth is an efficient tool for robustly summarizing the distribution of functional data and detecting potential magnitude and shape outliers. Commonly used functional data depth notions, such as the modified band depth and extremal depth, are estimated from pointwise depth for each observed functional observation. However, these techniques require calculating one single depth value for each functional observation, which may not be sufficient to characterize the distribution of the functional data and detect potential outliers. This article presents an innovative approach to make the best use of pointwise depth. We propose using the pointwise depth distribution for magnitude outlier visualization and the correlation between pairwise depth for shape outlier detection. Furthermore, a bootstrap-based testing procedure has been introduced for the correlation to test whether there is any shape outlier. The proposed univariate methods are then extended to bivariate functional data. The performance of the proposed methods is examined and compared to conventional outlier detection techniques by intensive simulation studies. In addition, the developed methods are applied to simulated solar energy datasets from a photovoltaic system. Results revealed that the proposed method offers superior detection performance over conventional techniques. These findings will benefit engineers and practitioners in monitoring photovoltaic systems by detecting unnoticed anomalies and outliers.

Identifying the underlying trajectory pattern in the spatial-temporal data analysis is a fundamental but challenging task. In this paper, we study the problem of simultaneously identifying temporal trends and spatial clusters of spatial-temporal trajectories. To achieve this goal, we propose a novel method named spatial clustered and sparse nonparametric regression ($��\mathrm{Scanner}�$). Our method leverages the B-spline model to fit the temporal data and penalty terms on spline coefficients to reveal the underlying spatial-temporal patterns. In particular, our method estimates the model by solving a doubly-penalized least square problem, in which we use a group sparse penalty for trend detection and a spanning tree-based fusion penalty for spatial cluster recovery. We also develop an algorithm based on the alternating direction method of multipliers (ADMM) algorithm to efficiently minimize the penalized least square loss. The statistical consistency properties of $��\mathrm{Scanner}�$ estimator are established in our work. In the end, we conduct thorough numerical experiments to verify our theoretical findings and validate that our method outperforms the existing competitive approaches.

This article deals with the measurement of groundwater arsenic ($��A�s�$) contamination. The focus is on using a proper index for the severity of contamination, rather than just using the proportion of observations above a threshold level. We specifically focus on the contamination severity index (CSI) proposed by Sen (2016. Sankhya B, 78B(2), 341–361.). An alternative estimator in contrast to the one given by Sen (2016. Sankhya B, 78B(2), 341–361.) is used here which is useful for a small number of observations. The data used is that collected by the British Geological Society and the Bangladesh Department of Public Health Engineering during 1997–2001. Their analysis was based on the simple proportion of the observations above a threshold level, whereas the CSI measure adequately takes into account the severity of the observations. It is emphasized in this article that the comparison of areas with average arsenic ($��A�s�$) levels to determine arsenic severity is not appropriate in general due to a large variation in the sample values due to the depth of wells. However, an alternative to the CSI proposed in Sen (2016. Sankhya B, 78B(2), 341–361.) has been given in this article that takes into account the depth of wells corresponding to the $��A�s�$ samples. This article also uses the bootstrap methodology in assessing the bias and standard errors of the estimators, and the corresponding bias-corrected and accelerated confidence intervals.

This article is devoted to the problem of detection of forest and nonforest areas on Earth images. We propose two statistical methods to tackle this problem: one based on multiple hypothesis testing with parametric distribution families, another one—on nonparametric tests. The parametric approach is novel in the literature and relevant to a larger class of problems—detection of natural objects, as well as anomaly detection. We develop mathematical background for each of the two methods, build self-sufficient detection algorithms using them and discuss practical aspects of their implementation. We also compare our algorithms with each other and with those from standard machine learning using satellite data.

The count data arise in many scientific areas. Our concerns here focus on spatial count responses with an excessive number of zeros and a set of available covariates. Estimating model parameters and selecting important covariates for spatial zero-inflated count models are both essential. Importantly, to alleviate deviations from model assumptions, we propose a spatial zero-inflated Poisson-like methodology to model this type of data, which relies only on assumptions for the first two moments of spatial count responses. We then design an effective iterative estimation procedure between the generalized estimating equation and the weighted least squares method to respectively estimate the regression coefficients and the variogram of the data model. Moreover, the stabilization of estimators is evaluated via a block jackknife technique. Furthermore, a distribution-free model selection criterion based on an estimate of the mean squared error of the estimated mean structure is proposed to select the best subset of covariates. The effectiveness of the proposed methodology is demonstrated by simulation studies under various scenarios, and a real dataset regarding the number of maternal deaths in Mozambique is analyzed for illustration.

In this article, we consider the notion of block-exogeneity and establish a characterization of it. We use this characterization to propose a procedure to test for block-exogeneity in a trivariate system. The proposed procedure has been applied to test the so-called galactic cosmic ray warming hypothesis. The galactic cosmic ray warming hypothesis suggests the existence of an indirect solar influence on Earth's climate. Our results seem to imply that this hypothesis does not hold. In particular, we find that the global temperature is block-exogenous with respect to both sunspot numbers (a measure of the solar activity) and galactic cosmic rays. This implies that the supposed indirect causal link from solar activity to temperature (through cosmic rays), postulated by the galactic cosmic ray warming hypothesis, does not appear to exist.

The heavy-tailed behavior of the generalized extreme-value distribution makes it a popular choice for modeling extreme events such as floods, droughts, heatwaves, wildfires and so forth. However, estimating the distribution's parameters using conventional maximum likelihood methods can be computationally intensive, even for moderate-sized datasets. To overcome this limitation, we propose a computationally efficient, likelihood-free estimation method utilizing a neural network. Through an extensive simulation study, we demonstrate that the proposed neural network-based method provides generalized extreme value distribution parameter estimates with comparable accuracy to the conventional maximum likelihood method but with a significant computational speedup. To account for estimation uncertainty, we utilize parametric bootstrapping, which is inherent in the trained network. Finally, we apply this method to 1000-year annual maximum temperature data from the Community Climate System Model version 3 across North America for three atmospheric concentrations: 289 ppm $��{�\text{CO}�}_{�2�}�$ (pre-industrial), 700 ppm $��{�\text{CO}�}_{�2�}�$ (future conditions), and 1400 ppm $��{�\text{CO}�}_{�2�}�$, and compare the results with those obtained using the maximum likelihood approach.

Big datasets are gathered daily from different remote sensing platforms. Recently, statistical co-kriging models, with the help of scalable techniques, have been able to combine such datasets by using spatially varying bias corrections. The associated Bayesian inference for these models is usually facilitated via Markov chain Monte Carlo (MCMC) methods which present (sometimes prohibitively) slow mixing and convergence because they require the simulation of high-dimensional random effect vectors from their posteriors given large datasets. To enable fast inference in big data spatial problems, we propose the recursive nearest neighbor co-kriging (RNNC) model. Based on this model, we develop two computationally efficient inferential procedures: (a) the collapsed RNNC which reduces the posterior sampling space by integrating out the latent processes, and (b) the conjugate RNNC, an MCMC free inference which significantly reduces the computational time without sacrificing prediction accuracy. An important highlight of conjugate RNNC is that it enables fast inference in massive multifidelity data sets by avoiding expensive integration algorithms. The efficient computational and good predictive performances of our proposed algorithms are demonstrated on benchmark examples and the analysis of the High-resolution Infrared Radiation Sounder data gathered from two NOAA polar orbiting satellites in which we managed to reduce the computational time from multiple hours to just a few minutes.

The ecological status of lakes is important for understanding an ecosystem's biodiversity as well as for service water quality and policies related to land use and agricultural run-off. If the status is weak, then decisions about management alternatives need to be made. We assess the value of information of lake monitoring in Finland, where lakes are abundant. With reasonable ecological values and restoration costs, the value of information analysis can be compared with the survey's costs. Data are worth gathering if the expected value from the data exceeds the costs. From existing data, we specify a hierarchical Bayesian spatial logistic regression model for the ecological status of lakes. We then rely on functional approximations and Laplace approximations to get closed-form expressions for the value of information of a sampling design. The case study contains thousands of lakes. The combinatorially difficult design problem is to wisely pick the right subset of lakes for data gathering. To solve this optimization problem, we study the performance of various heuristics: greedy forward algorithms, exchange algorithms and Bayesian optimization approaches. The value of information increases quickly when adding lakes to a small design but then flattens out. Good designs are usually composed of lakes that are difficult to manage, while also balancing a variety of covariates and geographic coverage. The designs achieved by forward selection are reasonably good, but we can outperform them with the more nuanced search algorithms. Statistical designs clearly outperform other designs selected according to simpler criteria.