Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Poisson regression”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Human NCR3 gene variants rs2736191 and rs11575837 alter longitudinal risk for development of pediatric malaria episodes and severe malarial anemia

Background: Plasmodium falciparum malaria is a leading cause of pediatric morbidity and mortality in holoendemic transmission areas. Severe malarial anemia [SMA, hemoglobin (Hb) < 5.0 g/dL in children] is the most common clinical manifestation of severe malaria in such regions. Although innate immune response genes are known to influence the development of SMA, the role of natural killer (NK) cells in malaria pathogenesis remains largely undefined. As such, we examined the impact of genetic variation in the gene encoding a primary NK cell receptor, natural cytotoxicity-triggering receptor 3 (NCR3), on the occurrence of malaria and SMA episodes over time. Methods: Susceptibility to malaria, SMA, and all-cause mortality was determined in carriers of NCR3 genetic variants (i.e., rs2736191:C > G and rs11575837:C > T) and their haplotypes. The prospective observational study was conducted over a 36 mos. follow-up period in a cohort of children (n = 1,515, aged 1.9–40 mos.) residing in a holoendemic P. falciparum transmission region, Siaya, Kenya. Results: Poisson regression modeling, controlling for anemia-promoting covariates, revealed a significantly increased risk of malaria in carriers of the homozygous mutant allele genotype (TT) for rs11575837 after multiple test correction [Incidence rate ratio (IRR) = 1.540, 95% CI = 1.114–2.129, P = 0.009]. Increased risk of SMA was observed for rs2736191 in children who inherited the CG genotype (IRR = 1.269, 95% CI = 1.009–1.597, P = 0.041) and in the additive model (presence of 1 or 2 copies) (IRR = 1.198, 95% CI = 1.030–1.393, P = 0.019), but was not significant after multiple test correction. Modeling of the haplotypes revealed that the CC haplotype had a significant additive effect for protection against SMA (i.e., reduced risk for development of SMA) after multiple test correction (IRR = 0.823, 95% CI = 0.711–0.952, P = 0.009). Although increased susceptibility to SMA was present in carriers of the GC haplotype (IRR = 1.276, 95% CI = 1.030–1.581, P = 0.026) with an additive effect (IRR = 1.182, 95% CI = 1.018–1.372, P = 0.029), the results did not remain significant after multiple test correction. None of the NCR3 genotypes or haplotypes were associated with all-cause mortality. Conclusions: Variation in NCR3 alters susceptibility to malaria and SMA during the acquisition of naturally-acquired malarial immunity. These results highlight the importance of NK cells in the innate immune response to malaria.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Disproportionate impacts of COVID-19 in a large US city

COVID-19 has disproportionately impacted individuals depending on where they live and work, and based on their race, ethnicity, and socioeconomic status. Studies have documented catastrophic disparities at critical points throughout the pandemic, but have not yet systematically tracked their severity through time. Using anonymized hospitalization data from March 11, 2020 to June 1, 2021 and fine-grain infection hospitalization rates, we estimate the time-varying burden of COVID-19 by age group and ZIP code in Austin, Texas. During this 15-month period, we estimate an overall 23.7% (95% CrI: 22.5–24.8%) infection rate and 29.4% (95% CrI: 28.0–31.0%) case reporting rate. Individuals over 65 were less likely to be infected than younger age groups (11.2% [95% CrI: 10.3–12.0%] vs 25.1% [95% CrI: 23.7–26.4%]), but more likely to be hospitalized (1,965 per 100,000 vs 376 per 100,000) and have their infections reported (53% [95% CrI: 49–57%] vs 28% [95% CrI: 27–30%]). We used a mixed effect poisson regression model to estimate disparities in infection and reporting rates as a function of social vulnerability. We compared ZIP codes ranking in the 75th percentile of vulnerability to those in the 25th percentile, and found that the more vulnerable communities had 2.5 (95% CrI: 2.0–3.0) times the infection rate and only 70% (95% CrI: 60%-82%) the reporting rate compared to the less vulnerable communities. Inequality persisted but declined significantly over the 15-month study period. Our results suggest that further public health efforts are needed to mitigate local COVID-19 disparities and that the CDC’s social vulnerability index may serve as a reliable predictor of risk on a local scale when surveillance data are limited.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2019

Radiation and Risk of Liver, Biliary Tract, and Pancreatic Cancers among Atomic Bomb Survivors in Hiroshima and Nagasaki: 1958–2009

The Life Span Study (LSS) of atomic bomb survivors has consistently demonstrated significant excess radiation-related risks of liver cancer since the first cancer incidence report. Here, we present updated information on radiation risks of liver, biliary tract and pancreatic cancers based on 11 additional years of follow-up since the last report, from 1958 to 2009. The current analyses used improved individual radiation doses and accounted for the effects of alcohol consumption, smoking and body mass index. The study participants included 105,444 LSS participants with known individual radiation dose and no known history of cancer at the start of follow-up. Cases were the first primary incident cancers of the liver (including intrahepatic bile duct), biliary tract (gallbladder and other and unspecified parts of biliary tract) or pancreas identified through linkage with population-based cancer registries in Hiroshima and Nagasaki. Poisson regression methods were used to estimate excess relative risks (ERRs) and excess absolute risks (EARs) associated with DS02R1 doses for liver (liver and biliary tract cancers) or pancreas (pancreatic cancer). We identified 2,016 incident liver cancer cases during the follow-up period. Radiation dose was significantly associated with liver cancer risk (ERR per Gy: 0.53, 95% CI: 0.23 to 0.89; EAR per 10,000 person-year Gy: 5.32, 95% CI: 2.49 to 8.51). There was no evidence for curvature in the radiation dose response (P=0.344). ERRs by age-at-exposure categories were significantly increased among those who were exposed at 0–9, 10–19 and 20–29 years, but not significantly increased after age 30 years, although there was no statistical evidence of heterogeneity in these ERRs (P = 0.378). The radiation ERRs were not affected by adjustment for smoking, alcohol consumption or body mass index. As in previously reported studies, radiation dose was not associated with risk of biliary tract cancer (ERR per Gy: –0.02, 95% CI: –0.25 to 0.30). Radiation dose was associated with a nonsignificant increase in pancreatic cancer risk (ERR per Gy: 0.38, 95% CI: <0 to 0.83). The increased risk was statistically significant among women (ERR per Gy: 0.70, 95% CI: 0.12 to 1.45), but not among men.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Nonsynonymous amino acid changes in the α-chain of complement component 5 influence longitudinal susceptibility to Plasmodium falciparum infections and severe malarial anemia in kenyan children

Background: Severe malarial anemia (SMA; Hb < 5.0 g/dl) is a leading cause of childhood morbidity and mortality in holoendemic Plasmodium falciparum transmission regions such as western Kenya. Methods: We investigated the relationship between two novel complement component 5 (C5) missense mutations [rs17216529:C>T, p(Val145Ile) and rs17610:C>T, p(Ser1310Asn)] and longitudinal outcomes of malaria in a cohort of Kenyan children (under 60 mos, n = 1,546). Molecular modeling was used to investigate the impact of the amino acid transitions on the C5 protein structure. Results: Prediction of the wild-type and mutant C5 protein structures did not reveal major changes to the overall structure. However, based on the position of the variants, subtle differences could impact on the stability of C5b. The influence of the C5 genotypes/haplotypes on the number of malaria and SMA episodes over 36 months was determined by Poisson regression modeling. Genotypic analyses revealed that inheritance of the homozygous mutant (TT) for rs17216529:C>T enhanced the risk for both malaria (incidence rate ratio, IRR = 1.144, 95%CI: 1.059–1.236, p = 0.001) and SMA (IRR = 1.627, 95%CI: 1.201–2.204, p = 0.002). In the haplotypic model, carriers of TC had increased risk of malaria (IRR = 1.068, 95%CI: 1.017–1.122, p = 0.009), while carriers of both wild-type alleles (CC) were protected against SMA (IRR = 0.679, 95%CI: 0.542–0.850, p = 0.001). Conclusion: Collectively, these findings show that the selected C5 missense mutations influence the longitudinal risk of malaria and SMA in immune-naïve children exposed to holoendemic P. falciparum transmission through a mechanism that remains to be defined.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

Poisson-response Tensor-on-Tensor Regression and Applications

We introduce Poisson-response tensor-on-tensor regression (PToTR), a novel regression framework designed to handle tensor responses composed element-wise of random Poisson-distributed counts. Tensors, or multi-dimensional arrays, composed of counts are common data in fields such as inter national relations, social networks, epidemiology, and medical imaging, where events occur across multiple dimensions like time, location, and dyads. PToTR accommodates such tensor responses alongside tensor covariates, providing a versatile tool for multi dimensional data analysis. We propose algorithms for maximum likelihood estimation under a canonical polyadic (CP) structure on the regression coefficient tensor that satisfy the positivity of Poisson parameters and then provide an initial theoretical error analysis for PToTR estimators. We also demonstrate the utility of PToTR through three concrete applications: longitudinal data analysis of the Integrated Crisis Early Warning System database, positron emission tomography (PET) image reconstruction, and change-point detection of communication patterns in longitudinal dyadic data. These applications highlight the versatility of PToTR in addressing complex, structured count data across various domains.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Analyzing count data with measurement error

In this article, we analyze observed count data such as the number of defects in a steel product where the observed counts are the true counts measured with errors. We account for the measurement error by using a measurement error model based on a latent lognormal (LLN) distribution. We consider making inference about a single population (e.g., from samples of a production lot) and a regression model (e.g., from runs of a designed experiment), where the measurement system properties are known, that is, the parameters of the LLN distribution are known. Then, we consider simultaneous inference for the single population and regression model as well as the measurement system. We demonstrate the proposed methodology with both simulated and real observed counts.

42 ENGINEERING↗
DOE OSTI2022

Analysis of overlapping count data

Counts of a specific characteristic were obtained within regions defined on an object that was manufactured in a proprietary setting. The count regions were altered during production and resulted in misaligned or overlapping count data. A closed-formula maximum likelihood estimator (MLE) of the new region means is derived using all of the available count data and an independent Poisson model. The MLE is shown to be preferable to estimators constructed using generalized linear models for the overlapping data setting. This closed-form estimator extends to over-dispersed overlapping count data as the quasi-MLE and also performs well with correlated overlapping count data. Standard errors for the estimator are approximated and are validated with a simulation study. Additionally, the methods are extended to overlapping multinomial data. Illustrative examples of the methods are provided throughout the paper and are reproducible with the supplemental R code. Additionally, proofs of the paper’s results are also included in the supplemental material.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

How much have adverse occupational health outcomes among construction workers improved over time? Evidence from 25 years of medical screening

Background Construction workers have always had a high risk of occupational illnesses. We used 25 years of data from a medical screening program serving older construction workers to determine how much health outcomes have improved over the past 60 years. Methods We investigated changes in relative risk for chest radiographs consistent with pneumoconiosis, COPD by spirometry, lung cancer mortality, and audiometry-assessed hearing impairment among workers participating in a medical screening program. Results were stratified by decade of first construction employment: before 1960, 1960–1969, 1970–1979, 1980–1989, and after 1990. Poisson and Cox regression analyses assessed relative risk by decade adjusted for age, sex, smoking, and years of construction trade work. Results Subjects were 94% male and, on average, 60 years old with 25 years of construction work. When compared to workers employed before 1960, those first employed after 1990 experienced the following reductions in model-adjusted relative risks: chronic obstructive pulmonary disease, 32%; all pneumoconiosis, 68%; parenchymal abnormalities, 35%; pleural abnormalities, 71%; hearing impairment, 20%; and lung cancer mortality, 48%. Risks started to decline in the 1960s with greatest reductions among workers first employed after 1970. Conclusions This study demonstrates the positive impact that adoption of occupational health protections have had over the past 60 years. The greatest risk reductions were observed for outcomes with strong regulatory and legal incentives to reduce exposures and associated risks, such as those associated with inhalation hazards (asbestos and silica), while lowest improvement was for hearing impairment, for which little regulatory enforcement and few prevention incentives have been adopted.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Foodborne illness outbreaks linked to unpasteurised milk and relationship to changes in state laws – United States, 1998–2018

Consumption of unpasteurised milk in the United States has presented a public health challenge for decades because of the increased risk of pathogen transmission causing illness outbreaks. We analysed Foodborne Disease Outbreak Surveillance System data to characterise unpasteurised milk outbreaks. Using Poisson and negative binomial regression, we compared the number of outbreaks and outbreak-associated illnesses between jurisdictions grouped by legal status of unpasteurised milk sale based on a May 2019 survey of state laws. During 2013–2018, 75 outbreaks with 675 illnesses occurred that were linked to unpasteurised milk; of these, 325 illnesses (48%) were among people aged 0–19 years. Of 74 single-state outbreaks, 58 (78%) occurred in states where the sale of unpasteurised milk was expressly allowed. Compared with jurisdictions where retail sales were prohibited (n = 24), those where sales were expressly allowed (n = 27) were estimated to have 3.2 (95% CI 1.4–7.6) times greater number of outbreaks; of these, jurisdictions where sale was allowed in retail stores (n = 14) had 3.6 (95% CI 1.3–9.6) times greater number of outbreaks compared with those where sale was allowed on-farm only (n = 13). This study supports findings of previously published reports indicating that state laws resulting in increased availability of unpasteurised milk are associated with more outbreak-associated illnesses and outbreaks.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Risk Ratio and Risk Difference Estimation in Case-cohort Studies

Background: In case-cohort studies with binary outcomes, ordinary logistic regression analyses have been widely used because of their computational simplicity. However, the resultant odds ratio estimates cannot be interpreted as relative risk measures unless the event rate is low. The risk ratio and risk difference are more favorable outcome measures that are directly interpreted as effect measures without the rare disease assumption. Methods: We provide pseudo-Poisson and pseudo-normal linear regression methods for estimating risk ratios and risk differences in analyses of case-cohort studies. These multivariate regression models are fitted by weighting the inverses of sampling probabilities. Also, the precisions of the risk ratio and risk difference estimators can be improved using auxiliary variable information, specifically by adapting the calibrated or estimated weights, which are readily measured on all samples from the whole cohort. Finally, we provide computational code in R (R Foundation for Statistical Computing, Vienna, Austria) that can easily perform these methods. Results: Through numerical analyses of artificially simulated data and the National Wilms Tumor Study data, accurate risk ratio and risk difference estimates were obtained using the pseudo-Poisson and pseudo-normal linear regression methods. Also, using the auxiliary variable information from the whole cohort, precisions of these estimators were markedly improved. Conclusion: The ordinary logistic regression analyses may provide uninterpretable effect measure estimates, and the risk ratio and risk difference estimation methods are effective alternative approaches for case-cohort studies. These methods are especially recommended under situations in which the event rate is not low.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Quantile regression-enriched event modeling framework for dropout analysis in high-temperature superconductor manufacturing

High-temperature superconductor (HTS) tapes have shown promising characteristics of high critical current, which are prerequisites for applications in high-field magnets. Due to the unstable growth conditions in the HTS manufacturing process, however, the frequent occurrences of dropouts in the critical current impede the consistent performance of HTS tapes. To manufacture HTS tapes with large scale, high yield, and uniform performance, it is essential to develop novel data analysis approaches for modeling the dropouts and identifying the related important process parameters. Conventional methods for modeling recurrent events, such as the point process, require the extraction of events from quality measurements. As the critical current is a continuous process, it may not comprehensively represent the drop patterns by transforming the time-series measurements into a set of events. Here, to solve this issue, we develop a novel quantile regression-enriched event modeling (QREM) framework that integrates the non-homogeneous Poisson process for modeling the occurrence of dropouts and the quantile regression for capturing the drop patterns. By incorporating the feature selection and regularization, the proposed framework identifies a set of significant process parameters that can potentially cause the dropouts of HTS tapes. The proposed method is tested on real HTS tapes produced using an advanced manufacturing process, successfully identifying important parameters that influence dropout events including the substrate temperature and voltage. The results demonstrate that the proposed QREM method outperforms the standard point process in predicting the occurrence of dropouts.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2026

Poisson Log-Normal Process for Count Data Prediction

Modeling count data is important in physics and other scientific disciplines, where measurements often involve discrete, non-negative quantities such as photon or neutrino detection events. Traditional parametric approaches can be trained to generate integer-count predictions but may struggle with capturing complex, non-linear dependencies often observed in the data. Gaussian process (GP) regression provides a robust non-parametric alternative to modeling continuous data; however, it cannot generate integer outputs. We propose the Poisson Log-Normal (PoLoN) process, a framework that employs GP to model Poisson log-rates. As in GP regression, our approach relies on the correlations between data points captured via GP kernel structure rather than explicit functional parameterizations. We demonstrate that the PoLoN predictive distribution is Poisson-LogNormal and provide an algorithm for optimizing kernel hyperparameters. Furthermore, we adapt the PoLoN approach to the problem of detecting weak localized signals superimposed on a smoothly varying background - a task of considerable interest in many areas of science and engineering. Our framework allows us to predict the strength, location and width of the detected signals. We evaluate PoLoN's performance using both synthetic and real-world datasets, including the open dataset from CERN which was used to detect the Higgs boson at the Large Hadron Collider. Our results indicate that the PoLoN process can be used as a non-parametric alternative for analyzing, predicting, and extracting signals from integer-valued data.

Saha, Anushka [Rutgers U., Piscataway]↗
DOE OSTI2024

Physics-Based Machine Learning Methods for U-235 Forensics Signatures

Signatures of low-intensity U-235 sources have been recently studied by utilizing a variety of machine learning (ML) classifiers using features derived from gamma spectral measurements collectedunder structured campaigns. Several ML classifiers, such as ensemble of tress and classification trees, revealed misleadingly-optimistic training error due to over-fitting, and furthermore,their performance is not directly relatable to the physical properties due to their data-driven, opaque designs. We present a regression-based ML method that first estimates the inverse distanceto the source and then utilizes a threshold to infer its presence, by representing the background as a source located at an infinite distance. For the inverse distance estimation, we study the ensembleof trees and Gaussian process regression methods, and a hyper parameter auto-tuning and selection method that employs five regression estimators. These methods avoid the over-fittingobserved in several ML classifiers, while providing the classification error nearly comparable to them based on independent test data. Their error is directly related to estimates of the inversephysical distance to source, and the precision of error determines the seperability property that determines the false alarm and missed detection rates. The property of monotonic decrease of thesource strength with increasing detector distance combined with Poisson distribution of measurements is utilized to analytically validate these methods by deriving the generalization equations ofunderlying regression methods.

Rao, Nageswara↗
DOE OSTI2024

Physics-Based Machine Learning Methods for U-235 Forensics Signatures

Signatures of low-intensity U-235 sources have been recently studied by utilizing a variety of machine learning (ML) classifiers using features derived from gamma spectral measurements collected under structured campaigns. Several ML classifiers, such as ensemble of tress and classification trees, revealed misleadingly-optimistic training error due to over-fitting, and furthermore, their performance is not directly relatable to the physical properties due to their data-driven, opaque designs. We present a regression-based ML method that first estimates the inverse distance to the source and then utilizes a threshold to infer its presence, by representing the background as a source located at an infinite distance. For the inverse distance estimation, we study the ensemble of trees and Gaussian process regression methods, and a hyper parameter auto-tuning and selection method that employs five regression estimators. These methods avoid the over-fitting observed in several ML classifiers, while providing the classification error nearly comparable to them based on independent test data. Their error is directly related to estimates of the inverse physical distance to source, and the precision of error determines the seperability property that determines the false alarm and missed detection rates. The property of monotonic decrease of the source strength with increasing detector distance combined with Poisson distribution of measurements is utilized to analytically validate these methods by deriving the generalization equations of underlying regression methods.

Rao, Nageswara↗
DOE OSTI2023

Simulation of Mechanical Fractionation of Chopped Whole-Plant Corn (WPC) Using Discrete Element Method (DEM)

Fractionating whole-plant corn (WPC) in a single-pass harvesting system requires studies on the WPC-to-equipment interaction for improved property control, as well as mechanical and air-driven separation processes compared to the traditional multi-pass grain and stover harvesting system. The discrete element method (DEM) technique has the potential to simulate WPC mechanical fractionation and support simulation-based design of WPC separation processes. In this study, methods to develop DEM particle models of WPC (kernel, cob, stalk, and husk) and their material properties for simulating mass fractionation using the ASABE standard mechanical shaker were proposed. Measurement was done on the axial dimensions (major, intermediate, and minor) and mass of each WPC type (mean sample size is 56), sampled from single-pass harvesting. Applying gaussian multivariate regression and bootstrapping re-sampling techniques, a DEM particle approximate to each WPC was developed. Sensitivity analysis of the DEM Young‘s modulus, Poisson‘s ratio, and interaction parameters of coefficient of restitution, coefficient of rolling friction, and coefficient of static friction on mass fraction was performed after 156 ASABE sieve-shaking DEM simulation runs, generated using Latin Hypercube Design (LHD) design of experiment (DOE) from 19 DEM material parameters. DEM simulation using Hertz-Mindlin with flexible bond contact laws and DOE optimized material properties successfully reproduced the mass fractions retained in ASABE sieves at 9.8% mean relative error and a coefficient of determination of R2 = 0.87. Here, the DEM methodology developed for mechanical WPC mass fractionation could be deployed to perform virtual design of feedstock handling equipment and performance analysis of mechanical fraction systems.

09 BIOMASS FUELS↗
DOE OSTI2023

Full-Field Strain Measurement Integrated with Two Dimension Regression Analysis to Evaluate the Bi-Modulus Elastic Properties of Isotropic and Transversely Isotropic Materials

Background: Measuring the physical properties of shale is critical for optimizing engineering activities such as geothermal energy generation and hydraulic fracturing. Shale is a transversely isotropic material. Furthermore, this material can also include micro and macro cracks at different locations and orientations that cause it to behave differently under tensile or compressive loading. Objective: In this work, a combined experimental–numerical approach is proposed to evaluate the bi-modulus elastic properties of isotropic and transversely isotropic materials. Methods: Full-field strain measurements for a circular disk under diametral compression are integrated with a regression analysis technique to evaluate the elastic properties of bi-modulus materials subjected to tensile and compressive loads using two loading configurations on the same specimen. Digital Image Correlation (DIC) is used to measure the full-field strains. Subsequently, in the case of an isotropic material, a linear least-squares approach is utilized to process the experimentally determined strains in conjunction with analytical expressions of the stress fields (in terms of far-field loading) to determine the elastic modulus E, the shear modulus G, and the Poisson’s ratio $v$. In the case of a transversely isotopic material, such as shale, a finite element model is implemented to determine the stress fields (again in terms of far-field loading), which is followed by repeating the previous regression analysis in an iterative process to estimate the elastic parameters. Results: The results show that the proposed technique successfully provides a complete set of elastic properties as a function of both the loading condition and the principal material directions. The technique is validated by measurements on a known isotropic material and then applied to determine the properties of shale. Conclusion: In this work, the proposed approach is successfully used to calculate the bi-modulus elastic response of poly(methyl meth- acrylate) (PMMA) and shale. As expected, PMMA exhibits an isotropic response with no bi-modulus effect, however, shale exhibits both transverse isotropy and a bi-modulus effect. Therefore, this approach holds promise for investigating the elastic properties of materials like rocks and fiber-reinforced composite laminates as functions of the principal material directions and the loading conditions.

42 ENGINEERING↗
DOE OSTI2024

Regularized Differentiation for Bioburden Density Estimation in Planetary Protection

In this paper, we propose and investigate the performance of two novel shrinkage estimators for bioburden density estimation in planetary protection. The estimators are based on the regularized differentiation of a cumulative count of colony forming units collected throughout the data collecting session or the life cycle of the entire mission. The regularized differentiation recasts the problem of bioburden density estimation as a linear least squares problem. The least squares problem is then solved through regularization techniques, such as truncated singular value decomposition and penalized least squares. The regularization is necessary to avoid noise amplification during the differentiation of noisy data. The two regularization estimators are compared with four other commonly used estimators to simultaneously evaluate the means of multivariable independent Poisson distributions: the maximum likelihood, noninformative Bayes estimator with Jeffreys prior, Empirical Bayes using conjugate gamma-Poisson model with gamma parameters selected by method of moments, and the Clevenson-Zidek estimator. It is shown through computer-simulated data that the regularized differentiation based on ridge regression has the smallest mean-squared error among all estimators. The analysis of shrinkage mechanism implemented by regularized differentiation is performed, and it is shown that the regularized differentiation amounts to performing a weighted averaging of all the samples. The weights are determined by the regularization parameter automatically selected by the L-curve technique. Since the method of least squares makes no distributional assumptions about the data, it presents an attractive technique for bioburden density estimation when there are concerns about the misspecification of the distributional model. The paper concludes with the analysis of the bioburden data collected during InSight mission and directions for future work.

97 - MATHEMATICS AND COMPUTING↗