检索结果-南通市图书馆

Open Journal of Statistics 2024年第3期14卷 207-227页

作者： mohamed m. shoukri Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Ontario Canada

Background: The signal-to-noise ratio (SNR) is recognized as an index of m.asurem.nts reproducibility. We derive the m.xim.m.likelihood estim.tors of SNR and discuss confidence interval construction on the difference between two correlated SNRs when the readings are from.bivariate norm.l and bivariate lognorm.l distribution. We use the Pearsons system.of curves to approxim.te the difference between the two estim.tes and use the bootstrap m.thods to validate the approxim.te distributions of the statistic of interest. m.thods: The paper uses the delta m.thod to find the first four central m.m.nts, and hence the skewness and kurtosis which are im.ortant in the determ.nation of the param.ters of the Pearsons distribution. Results: The approach is illustrated in two exam.les;one from.veterinary m.crobiology and food safety data and the other on data from.clinical m.dicine. We derived the four central m.m.nts of the target statistics, together with the bootstrap m.thod to evaluate the param.ters of Pearsons distribution. The fitted Pearsons curves of Types I and II were recom.ended based on the available data. The R-codes are also provided to be readily used by the readers.

关键词： Signal-to-Noise Ratio Bivariate Distributions Bootstrap m.thods Delta m.thod Pearson System.of Curves

Cautionary Rem.rks When Testing Agreem.nt between Two Raters for Continuous Scale m.asurem.nts: A Tutorial in Clinical Epidem.ology with Im.lem.ntation Using R

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Epidem.ology 2024年第1期14卷 56-74页

作者： mohamed m. shoukri Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Canada

Background: When continuous scale m.asurem.nts are available, agreem.nts between two m.asuring devices are assessed both graphically and analytically. In clinical investigations, Bland and Altm.n proposed plotting subject-wise differences between raters against subject-wise averages. In order to scientifically assess agreem.nt, Bartko recom.ended com.ining the graphical approach with the statistical analytic procedure suggested by Bradley and Blackwood. The advantage of using this approach is that it enables significance testing and sam.le size estim.tion. We noted that the direct use of the results of the regression is m.sleading and we provide a correction in this regard. m.thods: Graphical and linear m.dels are used to assess agreem.nts for continuous scale m.asurem.nts. We dem.nstrate that software linear regression results should not be readily used and we provided correct analytic procedures. The degrees of freedom.of the F-statistics are incorrectly reported, and we propose m.thods to overcom. this problem.by introducing the correct analytic form.of the F statistic. m.thods for sam.le size estim.tion using R-functions are also given. Results: We believe that the tutorial and the R-codes are useful tools for testing and estim.ting agreem.nt between two rating protocols for continuous scale m.asurem.nts. The interested reader m.y use the codes and apply them.to their available data when the issue of agreem.nt between two raters is the subject of interest.

关键词： Lim.ts of Agreem.nt Pitm.n and m.rgan Tests Test of Parallelism The Arcsine Variance Stabilizing Transform.tion Sam.le Size Estim.tion

Em.irical Bayesian Approach to Testing Hom.geneity of Several m.ans of Inflated Poisson Distributions (IPD)

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2023年第3期13卷 285-299页

作者： mohamed m. shoukri m.ha Aleid Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Canada Knowledge Translation Departm.nt Saudi National Institute of Health Riyadh Saudi Arabia

Objectives: We introduce a special form.of the Generalized Poisson Distribution. The distribution has one param.ter, yet it has a variance that is larger than the m.an a phenom.non known as “over dispersion”. We discuss potential applications of the distribution as a m.del of counts, and under the assum.tion of independence we will perform.statistical inference on the ratio of two m.ans, with generalization to testing the hom.geneity of several m.ans. m.thods: Bayesian m.thods depend on the choice of the prior distributions of the population param.ters. In this paper, we describe a Bayesian approach for estim.tion and inference on the param.ters of several independent Inflated Poisson (IPD) distributions with two possible priors, the first is the reciprocal of the square root of the Poisson param.ter and the other is a conjugate Gam.a prior. The param.ters of Gam.a distribution are estim.ted in the em.irical Bayesian fram.work using the m.xim.m.likelihood (m.) solution using nonlinear m.xed m.del (NLm.XED) in SAS. With these priors we construct the highest posterior confidence intervals on the ratio of two IPD param.ters and test the hom.geneity of several populations. Results: We encountered convergence problem.in estim.ting the hyperparam.ters of the posterior distribution using the NLm.XED. However, direct m.xim.zation of the predictive density produced solutions to the m.xim.m.likelihood equations. We apply the m.thodologies to RNA-SEQ read count data of gene expression values.

关键词： Distributions of Over-Dispersed Counts Lagrange Class of Distributions Knowledge Transfer Gam.a Prior Posterior Inference Wilson-Hilferty Transform.tion RNA_SEQ Read Counts Data

Quasi-Binom.al Regression m.del for the Analysis of Data with Extra-Binom.al Variation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2022年第1期12卷 1-14页

作者： mohamed m. shoukri m.ha m. Aleid Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Ontario Canada Departm.nt of Biostatistics Epidemiology and Scientific Computing King Faisal Specialist Hospital and Research Center Riyadh KSA

Objectives: Developing inference procedures on the quasi-binom.al distribution and the regression m.del. m.thods: Score testing and the m.thod of m.xim.m.likelihood for regression param.ters estim.tion. Data: Several exam.les are included, based on published data. Results: A quasi-binom.al m.del is used to m.del binary response data which exhibit extra-binom.al variation. A partial score test on the binom.al hypothesis versus the quasi-binom.al alternative is developed and illustrated on three data sets. The extended logit transform.tion on the binom.al param.ter is introduced and the large sam.le dispersion m.trix of the estim.ted param.ters is derived. The Nonlinear m.xed Procedure (NLm.XED) in SAS is shown to be very appropriate for the estim.tion of nonlinear regression.

关键词： Quasi-Binom.al Distribution Extra Binom.al Variations Score Test Quasi-Binom.al Regression m.del COVID-19 Case Fatality Data

Tem.oral Variations in m.rtality after Liver Transplantation: Retrospective Investigation of Potential Risk Factors Using Propensity Score

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Epidem.ology 2023年第4期13卷 250-259页

作者： Rana A. Alm.usa mohamed m. shoukri Departm.nt of Internal m.dicine King Saud Medical City Riyadh Saudi Arabia Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Ontario Canada

Objective: We aim.to detect over-tim. variations in m.rtality of liver transplant recipients stratified by the period of transplant. Since this is a retrospective investigation, bias reduction caused by possible confounding effects can be achieved by using propensity score weighting in a m.ltivariate logistic regression m.del. m.thods: m.dical charts of all adult liver transplant recipients (n = 250) who were transplanted in three periods 2005-2009, 2010-2014 and 2015-2019 were retrospectively reviewed. The following recipient factors were analyzed: recipients and donors’ ages, sex, renal im.airm.nt, body m.ss index (Bm.), presence of bacterial or viral infections, m.LD (m.del for end-stage diseases). m.ltivariate logistic m.del adjusted by Propensity Scores (PS) was used to identify the effect of the risk factors on m.rtality, and death within five years, in the targeted tim. fram.. Patient outcom.s are recorded as;(patient status = 1 if dead, or patient status = 0 if alive). Results: m.ld score, recipient age, and renal im.airm.nts were shown to be predictors of m.rtality in transplanted patients. m.ltivariate regression m.del was used to identify the significance of the specified risk factors, followed by pairwise com.arisons between periods. Pairwise com.arisons between periods using logistic regression weighted by the inverse propensity score, correcting for the possible confounding effect of m.asured covariates showed that the death rate is significantly reduced in subsequent periods as com.ared to the initial period. Conclusions: The clinical im.lications of these findings are the ability to stratify patients at high risk of posttransplant death by planning m.re intensive and accurate m.nagem.nt for them.

关键词： Liver Transplantation Age Body m.ss Index Renal Im.airm.nt Inverse Propensity Score Weighting

Quasi-Negative Binom.al: Properties, Param.tric Estim.tion, Regression m.del and Application to RNA-SEQ Data

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2022年第2期12卷 216-237页

作者： mohamed m. shoukri m.ha m. Aleid Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Canada Departm.nt of Biostatistics Epidemiology and Scientific Computing King Faisal Specialist Hospital and Research Center Riyadh Saudi Arabia

Background: The Poisson and the Negative Binom.al distributions are com.only used to m.del count data. The Poisson is characterized by the equality of m.an and variance whereas the Negative Binom.al has a variance larger than the m.an and therefore both m.dels are appropriate to m.del over-dispersed count data. Objectives: A new two-param.ter probability distribution called the Quasi-Negative Binom.al Distribution (QNBD) is being studied in this paper, generalizing the well-known negative binom.al distribution. This m.del turns out to be quite flexible for analyzing count data. Our m.in objectives are to estim.te the param.ters of the proposed distribution and to discuss its applicability to genetics data. As an application, we dem.nstrate that the QNBD regression representation is utilized to m.del genom.cs data sets. Results: The new distribution is shown to provide a good fit with respect to the “Akaike Inform.tion Criterion”, AIC, considered a m.asure of m.del goodness of fit. The proposed distribution m.y serve as a viable alternative to other distributions available in the literature for m.deling count data exhibiting overdispersion, arising in various fields of scientific investigation such as genom.cs and biom.dicine.

关键词： Queuing m.dels Overdispersion m.m.nt Estim.tors Delta m.thod Bootstrap m.xim.m.Likelihood Estim.tion Fisher’s Inform.tion Orthogonal Polynom.als Regression m.dels RNE-Seq Data

Investigating Trend in Cardiovascular Disease m.rtality and Its Association with Obesity in the Gulf Cooperative Council (GCC) Countries from.1990 to 2019

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Epidem.ology 2022年第3期12卷 221-230页

作者： Sarah Al-Gahtani Talal Abozaid Saad Al-Gahtani mohamed m. shoukri m.ha Aleid King Abdullah bin Abdulaziz University Hospital Riyadh Saudi Arabia King Khalid University Hospital Riyadh Saudi Arabia Im.m.m.ham.ad Ibn Saud Islam.c University Riyadh Saudi Arabia Schulich School of m.dicine and Dentistry University of Western Ontario London Ontario Canada King Faisal Specialist Hospital and Research Center Riyadh Saudi Arabia

Background: Cardiovascular diseases (CVDs) are the leading cause of death globally. An estim.ted 17.9 m.llion people died from.CVDs in 2019, representing 32% of all global deaths. Of these deaths, 85% were due to heart attack and stroke. Over three quarters of CVD deaths take place in low- and m.ddle-incom. countries. We have studied the pattern of m.rtality due to cardiovascular in the six countries of the Arabian Gulf and its association with obesity over the 29 years 1990 to 2019. m.thods: We used the linear m.xed effect m.dels to investigate the pattern of CVD m.rtality over the year 1990 to 2019, together with the pattern of change in one of the m.st im.ortant risk factors that is obesity, and its association with CVD m.rtality over the sam. period. Conclusions: Although there were fluctuations in the pattern of m.rtality and the prevalence of obesity over the specified period, there has been a steady decline in the per-100,000 num.er of deaths and the prevalence of obesity. However, there was a strong association between the two variables. From.the fitted m.dels we estim.ted that a one percent increase in obesity is associated with an average increase in cardiovascular deaths of 2.7 deaths per 100,000.

关键词： Cardiovascular Diseases Risk Factors Tim. Series Data Generalized Linear m.xed m.dels Predictive Analytics

Analysis of Length of Stay (LOS) Data from.the m.dical Records of Tertiary Care Hospital in Saudi Arabia for Five Diagnosis Related Groups: Application of Cox Prediction m.del

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2021年第1期11卷 99-112页

作者： Sara AL-Gahtani mohamed m. shoukri Departm.nt of Internal m.dicine King Fahd Medical City Riyadh Saudi Arabia Schulich School of m.dicine and Dentistry University of Western Ontario London Ontario Canada

Background: One of the m.in objectives of hospital m.nagem.nts is to control the length of stay (LOS). Successful control of LOS of inpatients will result in reduction in the cost of care, decrease in nosocom.al infec... 详细信息

Background: One of the main objectives of hospital managements is to control the length of stay (LOS). Successful control of LOS of inpatients will result in reduction in the cost of care, decrease in nosocomial infections, medication side effects, and better management of the limited number of available patients’ beds. The length of stay (LOS) is an important indicator of the efficiency of hospital management by improving the quality of treatment, and increased hospital profit with more efficient bed management. The purpose of this study was to model the distribution of LOS as a function of patient’s age, and the Diagnosis Related Groups (DRG), based on electronic medical records of a large tertiary care hospital. Materials and Methods: Information related to the research subjects were retrieved from a database of patients admitted to King Faisal Specialist Hospital and Research Center hospital in Riyadh, Saudi Arabia between January 2014 and December 2016. Subjects’ confidential information was masked from the investigators. The data analyses were reported visually, descriptively, and analytically using Cox proportional hazard regression model to predict the risk of long-stay when patients’ age and the DRG are considered as antecedent risk factors. Results: Predicting the risk of long stay depends significantly on the age at admission, and the DRG to which a patient belongs to. We demonstrated the validity of the Cox regression model for the available data as the proportionality assumption is shown to be satisfied. Two examples were presented to demonstrate the utility of the Cox model in this regard.

关键词： Diagnostic Related Groups K-m.ans Clustering In Hospital Length of Stay Cox Proportional Hazard m.dels Relative Risk Estim.tion

Inference Procedures on the Generalized Poisson Distribution from.m.ltiple Sam.les: Com.arisons with Nonparam.tric m.dels for Analysis of Covariance (ANCOVA) of Count Data

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2021年第3期11卷 420-436页

作者： m.ha Al-Eid mohamed m. shoukri Departm.nt of Biostatistics Epidemiology and Scientific Computing King Faisal Specialist Hospital and Research Center Riyadh Saudi Arabia Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry University of Western Ontario London Ontario Canada

Count data that exhibit over dispersion (variance of counts is larger than its m.an) are com.only analyzed using discrete distributions such as negative binom.al, Poisson inverse Gaussian and other m.dels. The Poisson is characterized by the equality of m.an and variance whereas the Negative Binom.al and the Poisson inverse Gaussian have variance larger than the m.an and therefore are m.re appropriate to m.del over-dispersed count data. As an alternative to these two m.dels, we shall use the generalized Poisson distribution for group com.arisons in the presence of m.ltiple covariates. This problem.is known as the ANCOVA and is solved for continuous data. Our objectives were to develop ANCOVA using the generalized Poisson distribution, and com.are its goodness of fit to that of the nonparam.tric Generalized Additive m.dels. We used real life data to show that the m.del perform. quite satisfactorily when com.ared to the nonparam.tric Generalized Additive m.dels.

关键词： Count Regression Over Dispersion Generalized Linear m.dels Analysis of Covariance Generalized Additive m.dels

On the Index of Repeatability: Estim.tion and Sam.le Size Requirem.nts

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2019年第4期9卷 530-541页

作者： m.ha Al-Eid mohamed m. shoukri Departm.nt of Biostatistics Epidemiology and Scientific Computing King Faisal Specialist Hospital and Research Center Riyadh KSA Departm.nt of Epidem.ology and Biostatistics Schulich School of Medicine and Dentistry London Ontario Canada

Background: Repeatability is a statem.nt on the m.gnitude of m.asurem.nt error. When biom.rkers are used for disease diagnoses, they should be m.asured accurately. Objectives: We derive an index of repeatability based on the ratio of two variance com.onents. Estim.tion of the index is derived from.the one-way Analysis of Variance table based on the one-way random.effects m.del. We estim.te the large sam.le variance of the estim.tor and assess its adequacy using bootstrap m.thods. An im.ortant requirem.nt for valid estim.tion of repeatability is the availability of m.ltiple observations on each subject taken by the sam. rater and under the sam. conditions. m.thods: We use the delta m.thod to derive the large sam.le variance of the estim.te of repeatability index. The question related to the num.er of required repeats per subjects is answered by two m.thods. In first m.thods we estim.te the num.er of repeats that m.nim.zes the variance of the estim.ted repeatability index, and the second determ.ne the num.er of repeats needed under cost-constraints. Results and Novel Contribution: The situation when the m.asurem.nts do not follow Gaussian distribution will be dealt with. It is shown that the required sam.le size is quite sensitive to the relative cost. We illustrate the m.thodologies on the Serum.Alanine-am.notransferase (ALT) available from.hospital registry data for sam.les of m.les and fem.les. Repeatability is higher am.ng fem.les in com.arison to m.les.

关键词： m.asurem.nts Errors Functions of Variance Com.onents Delta m.thod Lagrange m.ltiplier Bootstrap Technique