Statistics

The Mars Curiosity rover carries an instrument, ChemCam, designed to measure the composition of surface rocks and soil using laser-induced breakdown spectroscopy (LIBS). The measured spectra from this instrument must be analyzed to identify the component elements in the target sample, as well as their relative proportions. This process, which we call disaggregation, is complicated by so-called matrix effects, which describe nonlinear changes in the relative heights of emission lines as an unknown function of composition due to atomic interactions within the LIBS plasma. In this work we explore the use of the plasma physics code ATOMIC, developed at Los Alamos National Laboratory, for the disaggregation task. ATOMIC has recently been used to model LIBS spectra and can robustly reproduce matrix effects from first principles. The ability of ATOMIC to predict LIBS spectra presents an exciting opportunity to perform disaggregation in a manner not yet tried in the LIBS community, namely via Bayesian model calibration. However, using it directly to solve our inverse problem is computationally intractable due to the large parameter space and the computation time required to produce a single output. Therefore we also explore the use of emulators as a fast solution for this analysis. We discuss a proof of concept Gaussian process emulator for disaggregating two-element compounds of sodium and copper. The training and test datasets were simulated with ATOMIC using a Latin hypercube design. After testing the performance of the emulator, we successfully recover the composition of 25 test spectra with Bayesian model calibration.

With rapid transmission, the coronavirus disease 2019 (COVID-19) has led to over 2 million deaths worldwide, posing significant societal challenges. Understanding the spatial patterns of patient visits and detecting the local spreading events are crucial to controlling disease outbreaks. We analyze highly detailed COVID-19 contact tracing data collected from Seoul, which provides a unique opportunity to understand the mechanism of patient visit occurrence. Analyzing contact tracing data is challenging because patient visits show strong clustering patterns while clusters of events may have complex interaction behavior. To account for such behaviors, we develop a novel interaction Neyman-Scott process that regards the observed patient visit events as offsprings generated from a parent spreading event. Inference for such models is complicated since the likelihood involves intractable normalizing functions. To address this issue, we embed an auxiliary variable algorithm into our Markov chain Monte Carlo. We fit our model to several simulated and real data examples under different outbreak scenarios and show that our method can describe spatial patterns of patient visits well. We also provide visualization tools that can inform public health interventions for infectious diseases such as social distancing.

Mortality risk is a major concern to patients have just been discharged from the intensive care unit (ICU). Many studies have been directed to construct machine learning models to predict such risk. Although these models are highly accurate, they are less amenable to interpretation and clinicians are typically unable to gain further insights into the patients' health conditions and the underlying factors that influence their mortality risk. In this paper, we use patients' profiles extracted from the MIMIC-III clinical database to construct risk calculators based on different machine learning techniques such as logistic regression, decision trees, random forests and multilayer perceptrons. We perform an extensive benchmarking study that compares the most salient features as predicted by various methods. We observe a high degree of agreement across the considered machine learning methods; in particular, the cardiac surgery recovery unit, age, and blood urea nitrogen levels are commonly predicted to be the most salient features for determining patients' mortality risks. Our work has the potential for clinicians to interpret risk predictions.

Complex interconnections between information technology and digital control systems have significantly increased cybersecurity vulnerabilities in smart grids. Cyberattacks involving data integrity can be very disruptive because of their potential to compromise physical control by manipulating measurement data. This is especially true in large and complex electric networks that often rely on traditional intrusion detection systems focused on monitoring network traffic. In this paper, we develop an online detection algorithm to detect and localize covert attacks on smart grids. Using a network system model, we develop a theoretical framework by characterizing a covert attack on a generator bus in the network as sparse features in the state-estimation residuals. We leverage such sparsity via a regularized linear regression method to detect and localize covert attacks based on the regression coefficients. We conduct a comprehensive numerical study on both linear and nonlinear system models to validate our proposed method. The results show that our method outperforms conventional methods in both detection delay and localization accuracy.

In this article, we investigate the historical series of the total number of deaths per month in Brazil since 2015 using time series analysis techniques, in order to assess whether the COVID-19 pandemic caused any change in the series' generating mechanism. The results obtained so far indicate that there was no statistical significant impact.

Demand forecasting of hierarchical components is essential in manufacturing. However, its discussion in the machine-learning literature has been limited, and judgemental forecasts remain pervasive in the industry. Demand planners require easy-to-understand tools capable of delivering state-of-the-art results. This work presents an industry case of demand forecasting at one of the largest manufacturers of electronics in the world. It seeks to support practitioners with five contributions: (1) A benchmark of fourteen demand forecast methods applied to a relevant data set, (2) A data transformation technique yielding comparable results with state of the art, (3) An alternative to ARIMA based on matrix factorization, (4) A model selection technique based on topological data analysis for time series and (5) A novel data set. Organizations seeking to up-skill existing personnel and increase forecast accuracy will find value in this work.

Fire incidence is a big problem for every local government unit in the Philippines. The two most detrimental effects of fire incidence are economic loss and loss of life. To mitigate these losses, proper planning and implementation of control measures must be done. An essential aspect of planning and control measures is prediction of possible fire incidences. This study is conducted to analyze the historical data to create a forecasting model for the fire incidence in Davao City. Results of the analyses show that fire incidence has no trend or seasonality, and occurrences of fire are neither consistently increasing nor decreasing over time. Furthermore, the absence of seasonality in the data indicate that surge of fire incidence may occur at any time of the year. Therefore, fire prevention activities should be done all year round and not just during fire prevention month.

The healthcare sector in the U.S. is complex and is also a large sector that generates about 20% of the country's gross domestic product. Healthcare analytics has been used by researchers and practitioners to better understand the industry. In this paper, we examine and demonstrate the use of Beta regression models to study the utilization of brand name drugs in the U.S. to understand the variability of brand name drug utilization across different areas. The models are fitted to public datasets obtained from the Medicare & Medicaid Services and the Internal Revenue Service. Integrated Nested Laplace Approximation (INLA) is used to perform the inference. The numerical results show that Beta regression models can fit the brand name drug claim rates well and including spatial dependence improves the performance of the Beta regression models. Such models can be used to reflect the effect of prescription drug utilization when updating an insured's health risk in a risk scoring model.

One of the most common ways researchers compare survival outcomes across treatments when confounding is present is using Cox regression. This model is limited by its underlying assumption of proportional hazards; in some cases, substantial violations may occur. Here we present and compare approaches which attempt to address this issue, including Cox models with time-varying hazard ratios; parametric accelerated failure time models; Kaplan-Meier curves; and pseudo-observations. To adjust for differences between treatment groups, we use Inverse Probability of Treatment Weighting based on the propensity score. We examine clinically meaningful outcome measures that can be computed and directly compared across each method, namely, survival probability at time T, median survival, and restricted mean survival. We conduct simulation studies under a range of scenarios, and determine the biases, coverages, and standard errors of the Average Treatment Effects for each method. We then apply these approaches to two published observational studies of survival after cancer treatment. The first examines chemotherapy in sarcoma, where survival is very similar initially, but after two years the chemotherapy group shows a benefit. The other study is a comparison of surgical techniques for kidney cancer, where survival differences are attenuated over time.

The COVID-19 outbreak was initially reported in Wuhan, China, and it has been declared as a Public Health Emergency of International Concern (PHEIC) on 30 January 2020 by WHO. It has now spread to over 180 countries, and it has gradually evolved into a worldwide pandemic, endangering the state of global public health and becoming a serious threat to the global community. To combat and prevent the spread of the disease, all individuals should be well-informed of the rapidly changing state of COVID-19. To accomplish this objective, I have built a website to analyze and deliver the latest state of the disease and relevant analytical insights. The website is designed to cater to the general audience, and it aims to communicate insights through various straightforward and concise data visualizations that are supported by sound statistical methods, accurate data modeling, state-of-the-art natural language processing techniques, and reliable data sources. This paper discusses the major methodologies which are utilized to generate the insights displayed on the website, which include an automatic data ingestion pipeline, normalization techniques, moving average computation, ARIMA time-series forecasting, and logistic regression models. In addition, the paper highlights key discoveries that have been derived in regard to COVID-19 using the methodologies.