Article
Statistics & Probability
Tudor Manole, Abbas Khalili
Summary: Estimating the number of components in a finite mixture model is a challenging problem in statistics. The Group-Sort-Fuse (GSF) procedure proposes a new penalized likelihood approach to simultaneously estimate the order and mixing measure in multidimensional finite mixture models, achieving consistency in estimating the true mixture order and a convergence rate of n(-1/2) for parameter estimation. Its implementation in univariate and multivariate mixture models shows promising performance in both simulation studies and real data examples.
ANNALS OF STATISTICS
(2021)
Article
Mathematics
M. Kayid, Nassr S. Al-Maflehi
Summary: The quasi-Lindley distribution is a flexible model that is useful in various fields. This paper utilizes the expectation-maximization (EM) algorithm to estimate the parameters for both uncensored and right-censored data. Simulation studies demonstrate that the EM estimates perform better than maximum likelihood estimates (MLEs) in both scenarios. An illustrative example regarding the waiting times of a bank's customers is provided to compare the EM estimator with the MLE. The analysis of such data can be crucial in bank management.
JOURNAL OF MATHEMATICS
(2022)
Article
Mathematical & Computational Biology
Douwe Postmus, Francesco Pignatti, Hans L. Hillege, Tommi Tervonen
Summary: This article presents a simulated maximum likelihood procedure for fitting a Dirichlet population model directly to region-censored weights data and compares it to a naive approach. The results show that the proposed method outperforms the naive approach and can be applied with smaller sample sizes or fewer questions.
STATISTICS IN MEDICINE
(2022)
Article
Engineering, Civil
Abhishek Goyal, Alessia Flammini, Renato Morbidelli, Corrado Corradini, Rao S. Govindaraju
Summary: The impact of observations on the maximum likelihood estimates (MLE) of the Ks distribution parameters is evaluated in this study. Based on data from rainfall-runoff events, the results demonstrate the role of temporal variation of rainfall in resolving the Ks field for a rainfall event.
JOURNAL OF HYDROLOGY
(2023)
Article
Infectious Diseases
Alastair Ronald Main, Tariq Halasa, Ann Sofie Olesen, Louise Lohse, Thomas Bruun Rasmussen, Graham J. Belsham, Anette Boklund, Anette Botner, Lasse Engbo Christiansen
Summary: This study used data from controlled ASFV inoculations of pigs to assess transmission parameters and developed three models to analyze the transmission effects of different latency periods. The results showed that binary, piecewise-linear, and exponential models had good fit for latency periods of 4 or 5 days, while piecewise-linear and exponential models had the best fit for a latency period of 3 days. The estimated transmission parameters for ASFV II genotype through the air are similar to those for direct contact, and will be useful for future simulations of control measures against ASFV.
TRANSBOUNDARY AND EMERGING DISEASES
(2022)
Article
Political Science
Carlisle Rainey, Kelly McCaskey
Summary: The study introduces the penalized maximum likelihood (PML) estimator as a solution to the issue of substantial bias in logit model coefficients in small samples. Empirical evidence shows that the PML estimator offers a substantial improvement in small samples and noticeable improvement even in larger samples.
POLITICAL SCIENCE RESEARCH AND METHODS
(2021)
Article
Engineering, Multidisciplinary
Sadiah M. A. Aljeddani, M. A. Mohammed
Summary: In this study, a Power Generalized Weibull Distribution (PGWD) was introduced for estimating unknown variables in the progressing type-II-censoring method. The PGW parameters were estimated using the maximum likelihood, maximum product spacing, and Bayesian assessment approaches. The performance of the three techniques was compared using progressively type-II censored data and Monte Carlo Simulation. Bayesian estimation showed better performance in terms of mean squares of error and bias reduction for PGWD with a natural conjugate prior.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Mathematics, Interdisciplinary Applications
A. D. Barbour, Gesine Reinert
Summary: The article discusses the estimation issue of the rho parameter in P. Doreian's network disturbance model, pointing out that using maximum likelihood estimation method in dense graphs may result in bias and instability, and proposes a more intuitive estimator.
JOURNAL OF COMPLEX NETWORKS
(2021)
Article
Mathematics, Interdisciplinary Applications
Maria T. T. Barendse, Yves Rosseel
Summary: Pairwise maximum likelihood (PML) estimation is capable of accurately and efficiently estimating computationally intensive multilevel random intercept and random slope structural equation models (SEM) in the presence of discrete data and multiple latent variables.
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY
(2023)
Article
Engineering, Civil
Ildar R. Urazghildiiev
Summary: This article discusses the problem of using passive acoustics to estimate the position, speed, and heading angle of a moving source with underwater compact arrays. Maximum-likelihood (ML) estimators that utilize angle-of-arrival (AOA), time-difference-of-arrival (TDOA), and a combination of AOA/TDOA estimates are developed. The accuracy of the AOA-based, TDOA-based, and hybrid estimators is evaluated using Cramer-Rao bounds (CRB), statistical simulations, and in situ testing. Test results show that the proposed algorithms' practical accuracy heavily depends on deviations in speed and heading angles from their average values.
IEEE JOURNAL OF OCEANIC ENGINEERING
(2023)
Article
Public, Environmental & Occupational Health
Justin B. DeMonte, Anne M. Neilan, Matthew S. Loop, Andrea L. Ciaranello, Michael G. Hudgens
Summary: In this study, it was found that the crude incidence rate estimator in cross-sectional data is biased, while the maximum likelihood estimator is more accurate. The bias of the crude estimator tends to be much larger than the MLE in absolute value.
ANNALS OF EPIDEMIOLOGY
(2021)
Review
Biology
Paul Zaharias, Tandy Warnow
Summary: This article introduces some recent advances in highly accurate phylogeny estimation on large datasets, including divide-and-conquer techniques, methods for estimating species trees from multi-locus datasets and addressing heterogeneity, and methods for adding sequences into large gene trees or species trees.
PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES
(2022)
Article
Statistics & Probability
Jie Chen, Michael L. Stein
Summary: This study proposes a method to construct covariance functions with a hierarchical structure, which has efficient computational performance for large data scenarios. It is suitable for various random field computations such as sampling, kriging, and likelihood evaluation.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Article
Plant Sciences
Guanyu Zhu, G. S. V. Raghavan, Zhenfeng Li
Summary: This paper discusses the estimation problem of the moisture ratio model during a cantaloupe microwave drying process. A cantaloupe drying system based on image processing was designed, and the expression of the moisture ratio with regard to the shrinkage was built. An iterative evolution algorithm, based on the maximum likelihood principle (MLP-IE), was proposed to estimate the moisture ratio model. Furthermore, an improved iterative evolution algorithm (MLP-I-IE) was introduced to enhance the model fitting ability, by designing improved mutation strategy, scaling factor, and crossover rate. Both the MLP-IE algorithm and MLP-I-IE algorithm were effective, but the MLP-I-IE algorithm outperformed the MLP-IE algorithm in model estimation and validation.
Article
Computer Science, Artificial Intelligence
Yang Liu, Baoding Liu
Summary: This paper presents a new parameter estimation method in uncertain differential equations based on uncertain maximum likelihood estimation. Analytical formulae of the uncertain maximum likelihood estimators in special linear uncertain differential equations are given. Numerical examples are provided to illustrate this parameter estimation method.
Article
Audiology & Speech-Language Pathology
Yi Shen, Allison B. Kern, Virginia M. Richards
JOURNAL OF SPEECH LANGUAGE AND HEARING RESEARCH
(2019)
Article
Psychology
J. Devin McAuley, Yi Shen, Sarah Dec, Gary R. Kidd
ATTENTION PERCEPTION & PSYCHOPHYSICS
(2020)
Article
Acoustics
Yi Shen, Donghyeon Yun, Yi Liu
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA
(2020)
Article
Acoustics
Virginia M. Richards, Mariel Kazuko Tisby, Eli N. Suzuki-Gill, Yi Shen
Summary: Human listeners are able to perceptually recover the spectral profile of a complex sound when spectral components are distributed over time. The study used a cued informational masking paradigm to investigate the integration of spectral information over time. Results showed that reliable precursor cuing regarding the spectral content of the masker can significantly improve detectability of the target sound, but performance did not significantly improve with increased number of precursor bursts for many listeners.
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA
(2021)
Article
Psychology
J. Devin McAuley, Yi Shen, Toni Smith, Gary R. Kidd
Summary: Recent research showed that altering the rhythm of target speech affects target recognition, while altering background speech rhythm can enhance target recognition. The study found that the target-rhythm effect was present with all backgrounds, but the background-rhythm effect was mainly observed in same-sex background conditions.
ATTENTION PERCEPTION & PSYCHOPHYSICS
(2021)
Article
Neurosciences
Mackenzie L. Mills, Yi Shen, Robert H. Withnell
Summary: Research indicates that the growth of distortion product otoacoustic emission in humans is complex, mainly influenced by two-tone suppression and/or a two-component interaction. Additionally, a shift in the operating point may also be an important factor contributing to this complexity.
JARO-JOURNAL OF THE ASSOCIATION FOR RESEARCH IN OTOLARYNGOLOGY
(2021)
Article
Acoustics
Z. Ellen Peng, Sebastian Waz, Emily Buss, Yi Shen, Virginia Richards, Hari Bharadwaj, G. Christopher Stecker, Jordan A. Beim, Adam K. Bosen, Meredith D. Braza, Anna C. Diedesch, Claire M. Dorey, Andrew R. Dykstra, Frederick J. Gallun, Raymond L. Goldsworthy, Lincoln Gray, Eric C. Hoover, Antje Ihlefeld, Thomas Koelewijn, Judy G. Kopun, Juraj Mesik, Daniel E. Shub, Jonathan H. Venezia
Summary: Acoustics research is typically conducted in specialized laboratory settings, but remote testing allows for greater participant access and larger sample sizes in everyday environments. However, remote testing also has limitations such as reduced control over environmental conditions and potential inconsistencies in attentional state and response behaviors.
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA
(2022)
Article
Audiology & Speech-Language Pathology
Donghyeon Yun, Yi Shen, Jennifer J. Lentz
Summary: This study aims to validate the phase inversion technique for assessing the output signal-to-noise ratio (SNR) of digital hearing aids. By comparing the output SNRs computed via shadow filtering with those estimated using the phase inversion technique, it is found that the accuracy of the phase inversion technique depends on the test conditions. Therefore, caution should be exercised in using the phase inversion technique, and further evaluation of its validity is necessary.
AMERICAN JOURNAL OF AUDIOLOGY
(2023)
Article
Psychology, Multidisciplinary
Dylan V. Pearson, Yi Shen, J. Devin McAuley, Gary R. Kidd
Summary: Sensitivity to temporal properties of auditory patterns tends to be poorer in older listeners, contributing to their poorer speech understanding. This study examined sensitivity to speech rhythms in young and older normal-hearing subjects, finding that both rely on speech rhythms to generate temporal expectancies for upcoming events. However, older listeners do not exhibit lower thresholds for shortened gaps, indicating a change in speech-timing expectancies with age.
FRONTIERS IN PSYCHOLOGY
(2023)
Article
Acoustics
Yi Shen, Lauren Langley
Summary: This study estimated spectral weights in octave-frequency bands from 0.25 to 4 kHz for speech-in-noise recognition using two types of sentence materials. The masking noise was either unmodulated or sinusoidally amplitude-modulated at 8 Hz. The results showed that the estimated spectral weights were consistent across two test sessions and similar for the two types of sentence materials. Amplitude modulation of the masker affected the weights at different frequencies, suggesting a possible shift in spectral weights for temporally fluctuating maskers.
JASA EXPRESS LETTERS
(2023)
Article
Audiology & Speech-Language Pathology
Dylan Pearson, Yi Shen, J. Devin McAuley, Gary R. Kidd
Summary: Understanding continuous speech with competing background sounds is challenging, especially for older adults. Temporal regularity (rhythm) is a stimulus property that may help listeners understand target material. This study replicates a target rhythm effect in young adults and investigates age-related changes in the effect of rhythm on recognition. The results show that temporal regularity improves identification accuracy in young listeners and older listeners also demonstrate a target rhythm effect but receive less benefit from regularity.
Proceedings Paper
Acoustics
Zhuohuang Zhang, Donald S. Williamson, Yi Shen
Summary: This paper investigates the importance of estimating phase across different spectral regions and proposes a hybrid deep-learning framework for handling phase differently across the spectrum. The study suggests that estimating phase at lower-frequency bands is important for speech quality, and the proposed hybrid-net achieves superior performance.
Article
Acoustics
Donghyeon Yun, Yi Shen, Zhuohuang Zhang
Summary: Personal hearing devices can be fine-tuned through self-adjustment by users. Two self-adjustment procedures were developed to explore a multi-dimensional parameter space rapidly. Monte Carlo simulations and behavioral evaluations showed that these procedures are effective and reliable.
JOURNAL OF THE ACOUSTICAL SOCIETY OF KOREA
(2022)
Meeting Abstract
Clinical Neurology
Yi Shen, Jian Wu
Proceedings Paper
Acoustics
Zhuohuang Zhang, Yi Shen, Donald S. Williamson
2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2019)