Consistent Estimates Research Articles

Solving the Price-Setting Newsvendor Problem with Parametric Operational Data Analytics (ODA)

We study the data-integrated, price-setting newsvendor problem in which the price–demand relationship is described by some parametric model with unknown parameters. We develop the operational data analytics (ODA) formulation of this problem that features a data-integration model and a validation model. The data-integration model consists of a class of functions called the operational statistics. Each operational statistic maps the available data to the ordering decision. The validation model finds, among the set of candidate operational statistics, the ordering decision that leads to the highest actual profit, which is unknown because of the unknown demand parameters. This ODA framework leads to a consistent estimate of the profit function with which we optimize the pricing decision. The derived quantity and price decisions demonstrate robust profit performance even when the sample size is very small in relation to the demand variability. Compared with the conventional approach with which the unknown parameters are estimated and then the decisions are optimized, the ODA framework produces significantly superior performance in the mean, standard deviation, and minimum of the profit, suggesting the robustness of the ODA solution especially in the small-sample regime. This paper was accepted by David Simchi-Levi, operations management. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2021.02227 .

Abstract 4135517: Echocardiographic Biomarkers Enhance Mortality Predictions in Cirrhosis Patients

Introduction: Cirrhosis, a late-stage liver disease, is associated with numerous systemic complications, making long-term mortality prediction challenging. Traditional models like Child-Pugh and MELD are primarily designed for short-term (3-month) mortality. This study evaluates the impact of incorporating echocardiographic (echo) variables in predicting long-term mortality in cirrhosis patients. Methods: A retrospective cohort study was conducted using data from 60 patients with clinically diagnosed cirrhosis from Hippokration Hospital, Greece, spanning 2018 to 2022. Patients were followed for up to 52 months. Exclusion criteria included pre-existing cardiovascular diseases and significant co-morbidities. Patients were categorized into survived to end of study (n=31) and died during study (n=29) groups. Outcomes were analyzed using multivariate logistic regression and Cox proportional hazards models, and include parameters for age, blood pressure, lab-based biomarkers (potassium, pro-brain natriuretic peptide, hemoglobin, platelet count, aspartate transaminase, gamma-glutamyl transferase, uric acid), and echo/ecg measurements (QT interval, left ventricular (LV) ejection fraction, LV global longitudinal strain (GLS), left atrial GLS, annulus velocities, and E/A ratio). Results: The multivariate logistic regression model (MELD+) that included echo variables showed a 10% improvement in area under the curve (AUC: 0.901) compared to the MELD model alone (AUC: 0.817). MELD+ improved risk classification accuracy by nearly 10% with a conservative threshold. The Cox proportional hazards model with MELD+ also demonstrated superior performance (AUC: 0.923; C-index: 0.875; Brier Score: 0.097) compared to the MELD model alone (AUC: 0.892; C-index: 0.831; Brier Score: 0.134). The MELD+ model showed less variation in hazard ratio confidence intervals, indicating more consistent risk estimation. Conclusion: Incorporating echocardiographic measurements enhances the accuracy of long-term mortality predictions in cirrhosis patients. These findings suggest that cardiac imaging should be considered in risk assessments for cirrhosis patients, particularly for predicting survival beyond three months.

Abstract 4146312: Global Burden and Trend of Atrial Fibrillation and Flutter in the 27 European Union Countries from 1990-2021: A Systematic analysis for the Global Burden of Disease Study 2021

Introduction: Atrial Fibrillation (Afib) and Flutter ranks as the fifth leading cause of death among all cardiovascular diseases (CVD). Given the scarcity of consistent previous estimates, this study is the first to assess the burden of Afib and Flutter in the European Union (EU) over the past three decades, including the initial two years of the COVID-19 pandemic. Methods: Using global burden of disease study 2021 meta tool, we estimated prevalence, incidence, deaths, disability adjusted life years (DALYs), years lived with disability (YLDs) due to Afib and Flutter by age, sex, year and location across the 27 EU countries from 1990-2021. Results: The total number of prevalence rose from 5 million (95% uncertainty interval: 3.9-6.4 million) in 1990 to 8.6 million (7.2-10.3 million) in 2021. The total percentage of change (TPC) in deaths increased by 130% (110%-143%), and YLDs by 69% (56%-84%) from 1990-2021. Austria saw the highest increase in age-standardized incidence rates (ASIR) at 83%, followed by Czechia at 50%. Sweden experienced the largest rise in mortality rates (ASMR) at 93%, with Estonia at 35%, and in YLD rates (ASYLDR), Austria led with a 90% increase followed by Sweden at 49%. In terms of age, individuals aged 55 and older recorded the highest death toll at 70,269 (57,793-77,019) and the highest incidence at 607,960 (397,686-868,122) in 2021. Regarding gender, males showed an increased TPC in overall burden compared to females, with males observing an increase and females a decreasing trend in ASIR (5% vs -7%), ASMR (7% vs -4%), and ASYLDR (7% vs -5%) from 1990-2021. Conclusion: Deaths due to Afib and Flutter accounted for 4.26% of all CVD in EU. The escalating prevalence and mortality rates of Afib and Flutter across the EU highlight an urgent need for comprehensive healthcare strategies. Effective management should focus on bolstering preventive measures, advancing diagnostic techniques, and enhancing patient care frameworks, particularly for the aging demographic most at risk. Strategic collaboration across sectors, including innovative public health initiatives and policy reinforcement, is essential to curb this growing trend and safeguard public health.

FITTING DYNAMICALLY CONSISTENT FORWARD RATE CURVES: ALGORITHM AND COMPARISON

This paper introduces a new methodology for estimating dynamically consistent forward rate curves, which are essential for obtaining arbitrage-free valuation and risk management models. Using U.S. Treasury data from January 2013 to June 2023, we fit dynamically consistent forward rates curves and test the goodness of fit between theoretical and observed prices. We compare the traditional static forward rate curve parameterizations (including Nelson–Siegel, Svensson, and cubic splines) with our new dynamically consistent methodology. The dynamically consistent forward rate curve estimates generally outperform the traditional static methods in reproducing market prices.

A General Framework for the Multiple Nonparametric Behrens-Fisher Problem With Dependent Replicates.

In many trials and experiments, subjects are not only observed once but multiple times, resulting in a cluster of possibly correlated observations (e.g., brain regions per patient). Observations often do not fulfill model assumptions of mixed models and require the use of nonparametric methods. In this article, we develop and present a purely nonparametric rank-based procedure that flexibly allows the unbiased and consistent estimation of the Wilcoxon-Mann-Whitney effect in clustered data designs. Compared with existing methods, we allow flexible weights to be used in effect estimation. Additionally, we develop global and multiple contrast test procedures to test null hypotheses formulated regarding the generalized Mann-Whitney effects and for the computation of range-preserving simultaneous confidence intervals in a unified way. Extensive simulation studies show that these methods control the type-I error rate well and have reasonable power to detect alternatives in various situations.

Data‐driven estimation for multithreshold accelerated failure time model

AbstractThis article develops a novel estimation framework for the multithreshold accelerated failure time model, which has distinct linear forms within different subdomains. One major challenge is to determine the number of threshold effects. We first show the selection consistency of a modified Bayesian information criterion under mild conditions. It is useful sometimes but heavily depends on the penalization magnitude, which usually varies from the model configuration and data distribution. To address this issue, we leverage a cross‐validation criterion alongside an order‐preserved sample‐splitting scheme to yield a consistent estimation. The new criterion is completely data driven without additional parameters and thus robust to model setting and data distributions. The asymptotic properties for the parameter estimates are also carefully established. Additionally, we propose an efficient score‐type test to examine the existence of threshold effects. The new statistic is free of estimating any potential threshold effects and is thus suitable for multithreshold scenarios. Numerical experiments validate the reliable finite‐sample performance of our methodologies, which corroborates the theoretical results.

The Impact of Non‐Monolithic Semiconductor Capacitance on Organic Electrochemical Transistors Performance and Design

AbstractThe existing device models for organic electrochemical transistors (OECTs) fail to provide any device design guidelines for optimized performance parameters such as transconductance that are pivotal for the applications OECTs in sensing. Moreover, the current models are based on the questionable assumption of a homogenous organic semiconductor layer, and all predict a linear behavior of the resistance with the OECT channel length. Consequently, the experimentally observed nonlinear resistance behavior in OECTs has been overlooked thus far. Here, an OECT device model is developed that accurately describes the nonlinear behavior of the OECT channel resistance and offers the first guidelines for maximizing transconductance. The model is inherently nonlinear and the nonlinearity stem from the non‐monolithic capacitance of the organic semiconductor layer. Moreover, the model provides a consistent and reliable estimations for the contact resistance in OECTs. The success of the model in accurately describing and providing predictions of the OECT operation by relating the device's geometrical parameters with electrochemical parameters of the semiconductor layer paves the way toward unlocking OECT potentials in diverse applications, from biosensing to neuromorphic computing and flexible electronics.

Residual extropy in ranked set sampling: Properties, comparative analysis, and estimation

Ranked set sampling (RSS) has gained significant significance in numerous practical domains, such as environmental and ecological studies, and statistical genetics. This article focuses on investigating the concept of extropy as a measure of residual uncertainty in RSS. It is devoted to discussing various monotone properties and characterization results associated with the residual extropy of RSS. Additionally, a comparative analysis is conducted between the residual extropy of RSS and its counterpart in simple random sampling. Optimal minima and maxima values for the residual extropy of RSS are determined. To further contribute to the field, a consistent estimator for the residual extropy of RSS is proposed. The effectiveness of this estimator is demonstrated through three illustrative examples, highlighting its performance in practical scenarios.

On the Relationship Among Linear Type Traits and Functional Longevity in the Italian Mediterranean Buffalo Using a Weibull Proportional Hazards Model

Weibull proportional hazards model was used to identify the non-genetic effects that affect length of productive life (LPL) of the Italian Mediterranean Buffalo (IMB). Data were provided by the Italian National Association of Buffalo Breeders (ANASB) and included records of reproductive, productive, and linear type traits from 59,943 buffalo with first calving from 2002 to 2019. Data were divided into 4 geographical regions to determine if the relationship between the investigated effects and LPL varied by region. LPL was defined as the number of days from the first calving to culling for those buffaloes that were culled (uncensored) or to the date of the last test-day for those that are still alive (censored). The Weibull model included time-dependent effects of herd-year-season of calving, parity and stage of lactation, production as within-herd deviations, as well as time-independent effects of age at first calving, year of birth, type of reproduction (natural mating vs artificial insemination), classifier and ten linear type traits. The average duration of productive life was 1,604 days. Both year of birth and production level had a significant effect on culling risk. Moreover, culling risk decreased linearly across parities. The reference point for age at first calving relative to other classes was set at 35 months. A significant relationship between linear type traits and relative culling risk was also observed, being highest for buffaloes with the lowest linear scores. The results show that the Weibull model provides consistent and robust risk estimates. Therefore, this model would be recommended for future implementation of the first genetic evaluation of LPL in the Italian Mediterranean Buffalo.

Sieve estimation of the accelerated mean model based on panel count data

Panel count data are gathered when subjects are examined at discrete times during a study, and only the number of recurrent events occurring before each examination time is recorded. We consider a semiparametric accelerated mean model for panel count data in which the effect of the covariates is to transform the time scale of the baseline mean function. Semiparametric inference for the model is inherently challenging because the finite-dimensional regression parameters appear in the argument of the (infinite-dimensional) functional parameter, i.e., the baseline mean function, leading to the phenomenon of bundled parameters. We propose sieve pseudolikelihood and likelihood methods to construct the random criterion function for estimating the model parameters. An inexact block coordinate ascent algorithm is used to obtain these estimators. We establish the consistency and rate of convergence of the proposed estimators, as well as the asymptotic normality of the estimators of the regression parameters. Novel consistent estimators of the asymptotic covariances of the estimated regression parameters are derived by leveraging the counting process associated with the examination times. Comprehensive simulation studies demonstrate that the optimization algorithm is much less sensitive to the initial values than the Newton–Raphson method. The proposed estimators perform well for practical sample sizes, and are more efficient than existing methods. An example based on real data shows that due to this efficiency gain, the proposed method is better able to detect the significance of practically meaningful covariates than an existing method.

Asymptotic normality of estimators for all parameters in the Vasicek model by discrete observations

We continue to study the joint estimation of all three parameters of the Vasicek model, based on equidistant observations of a trajectory. We consider strongly consistent estimators previously constructed in Prykhodko and Ralchenko (2022) and establish their joint asymptotic normality. The theoretical results are illustrated by a numerical study that confirms our theoretical findings.

On the estimation of bivariate conditional transition rates

Recent literature has found conditional transition rates to be a useful tool for avoiding Markov assumptions in multi-state models. While the estimation of univariate conditional transition rates has been extensively studied, the intertemporal dependencies captured in the bivariate conditional transition rates still require a consistent estimator. We provide an estimator that is suitable for censored data and emphasize the connection to the rich theory of the estimation of bivariate survival functions. Bivariate conditional transition rates are necessary for various applications in the survival context but especially in the calculation of moments in life insurance mathematics.

EXPRESS: Addressing Endogeneity Using a Two-stage Copula Generated Regressor Approach

The ubiquitous presence of endogenous regressors presents a significant challenge when drawing causal inference using observational data. The classical econometric method used to handle regressor endogeneity requires IVs that must satisfy the stringent condition of exclusion restriction, rendering it unfeasible in many settings. Herein, we propose a new IV-free method that uses copulas to address the endogeneity problem. Existing copula correction methods require nonnormal endogenous regressors: normally or nearly normally distributed endogenous regressors cause model non-identification or significant finite-sample bias. Furthermore, existing copula control function methods presume the independence of exogenous regressors and the copula control function. While maintaining the Gaussian copula regressor-error dependence structure, our generalized two-stage copula endogeneity correction (2sCOPE) method simultaneously relaxes the two key identification requirements. Under the Gaussian copula dependence structure, we prove that 2sCOPE yields consistent causal-effect estimates with correlated endogenous and exogenous regressors as well as normally distributed endogenous regressors. In addition to relaxing the identification requirements, 2sCOPE has superior finite-sample performance and addresses the significant finite-sample bias problem due to insufficient regressor nonnormality. Moreover, 2sCOPE employs generated regressors derived from existing regressors to control for endogeneity, and can thus considerably increase the ease and broaden the applicability of IV-free methods for handling regressor endogeneity. We further demonstrate 2sCOPE’s performance using simulation studies and illustrate its use in an empirical application.

Utility of polygenic risk score for prediction of dilated cardiomyopathy and modulation of severity of LV dysfunction among cases

Abstract Background Dilated cardiomyopathy (DCM) is a predominant cause of heart failure and a leading indication for cardiac transplantation. While rare genetic variants are established as causative factors in familial forms of the disease, recent genome-wide association studies (GWAS) have also underscored the significant contribution of common genetic variation to DCM. Polygenic scores (PGS) constructed from GWAS have demonstrated promise in risk stratification for a range of common diseases and may similarly have value for risk prediction in DCM. However, data regarding the clinical utility of PGS in DCM remains sparse. Purpose We aimed to evaluate the utility of PGS for prediction of DCM and its association with left ventricular ejection fraction (LVEF) and family history. We further aimed to assess the contribution of PGS in DCM patients with established causative rare variants (genotype-positive) and those without (genotype-negative). Methods We used a previously-constructed PGS, which was developed from a GWAS meta-analysis (8387 DCM cases and 939161 controls) and multi-trait analysis (MTAG) with cardiac MRI traits. In the present analysis, we calculated PGS for a newly assembled cohort of 978 DCM cases from a major university medical centre and 7207 controls from a national register. To evaluate the clinical utility of PGS, we built logistic regression models and assessed the association between PGS and DCM status. Subsequently, we performed subgroup analyses, including: i) individuals of European ancestry, ii) non-European ancestry, iii) males, iv) females, v) genotype-positive participants(n=193), and vi) genotype-negative participants (n=294). We then used linear and logistic regression models, respectively, to assess the associations of PGS with LVEF and documented family history of DCM. Results PGS demonstrated a significant enrichment among DCM cases compared to controls (OR per SD 1.93, P = 9.47E-68). This association persisted with consistent effect estimates - ranging from OR 1.5 to 2.2 - across all examined subgroups and demonstrated sufficient statistical significance (all P<1.58E-06) (Figure 1). Genotype-negative cases had a significantly higher PGS compared to genotype-positive individuals (P = 0.0015), although PGS was significantly enriched in both groups as compared to controls. Furthermore, among DCM cases, higher PGS was significantly associated with lower LVEF at first presentation (P=0.03, beta=-0.5% per SD) (Figure 2). There was no significant association found between PGS and family history. Conclusions PGS is strongly associated with risk of DCM, and may modulate the severity of LV dysfunction among DCM cases. Polygenic burden contributes to DCM risk in both genotype-positive and negative cases, although the contribution is stronger in patients without known causative rare variants. Overall, our results suggest a potential clinical applicability for PGS in DCM risk stratification.Figure 1.Association of PGS with DCMFigure 2.Association of PGS with LVEF

3D distortion-free, reduced FOV diffusion-prepared gradient echo at 3 T.

To develop a 3D distortion-free reduced-FOV diffusion-prepared gradient-echo sequence and demonstrate its application in vivo for diffusion imaging of the spinal cord in healthy volunteers. A 3D multi-shot reduced-FOV diffusion-prepared gradient-echo acquisition is achieved using a slice-selective tip-down pulse in the phase-encoding direction in the diffusion preparation, combined with magnitude stabilizers, centric k-space encoding, and 2D phase navigators to correct for intershot phase errors. The accuracy of the ADC values obtained using the proposed approach was evaluated in a diffusion phantom and compared to the tabulated reference ADC values and to the ADC values obtained using a standard spin echo diffusion-weighted single-shot EPI sequence (DW-SS-EPI). Five healthy volunteers were scanned at 3 T using the proposed sequence, DW-SS-EPI, and a clinical diffusion-weighted multi-shot readout-segmented EPI sequence (RESOLVE) for cervical spinal cord imaging. Image quality, perceived SNR, and image distortion were assessed by two expert radiologists. ADC maps were calculated, and ADC values obtained with the proposed sequence were compared to those obtained using DW-SS-EPI and RESOLVE. Consistent ADC estimates were measured in the diffusion phantom with the proposed sequence and the conventional DW-SS-EPI sequence, and the ADC values were in close agreement with the reference values provided by the manufacturer of the phantom. In vivo, the proposed sequence demonstrated improved image quality, improved perceived SNR, and reduced perceived distortion compared to DW-SS-EPI, whereas all measures were comparable against RESOLVE. There were no significant differences in ADC values estimated in vivo for each of the sequences. 3D distortion-free diffusion-prepared imaging can be achieved using the proposed sequence.

Direct Estimation for Commonly Used Pattern-Mixture Models in Clinical Trials

Pattern-mixture models have received increasing attention as they are commonly used to assess treatment effects in primary or sensitivity analyses for clinical trials with nonignorable missing data. Pattern-mixture models have traditionally been implemented using multiple imputation, where the variance estimation may be a challenge because the Rubin’s approach of combining between- and within-imputation variance may not provide consistent variance estimation while bootstrap methods may be time-consuming. Direct likelihood-based approaches have been proposed in the literature and implemented for some pattern-mixture models, but the assumptions are sometimes restrictive and the theoretical framework is fragile. In this article, we propose an analytical framework for an efficient direct likelihood estimation method for commonly used pattern-mixture models corresponding to return-to-baseline, jump-to-reference, placebo washout, and retrieved dropout imputations. A parsimonious tipping point analysis is also discussed and implemented. Results from simulation studies demonstrate that the proposed methods provide consistent estimators. We further illustrate the utility of the proposed methods using data from a clinical trial evaluating a treatment for type 2 diabetes.

Two-stage pseudo maximum likelihood estimation of semiparametric copula-based regression models for semi-competing risks data.

We propose a two-stage estimation procedure for a copula-based model with semi-competing risks data, where the non-terminal event is subject to dependent censoring by the terminal event, and both events are subject to independent censoring. With a copula-based model, the marginal survival functions of individual event times are specified by semiparametric transformation models, and the dependence between the bivariate event times is specified by a parametric copula function. For the estimation procedure, in the first stage, the parameters associated with the marginal of the terminal event are estimated using only the corresponding observed outcomes, and in the second stage, the marginal parameters for the non-terminal event time and the copula parameter are estimated together via maximizing a pseudo-likelihood function based on the joint distribution of the bivariate event times. We derived the asymptotic properties of the proposed estimator and provided an analytic variance estimator for inference. Through simulation studies, we showed that our approach leads to consistent estimates with less computational cost and more robustness than the one-stage procedure developed in Chen YH (Lifetime Data Anal 18:36-57, 2012), where all parameters were estimated simultaneously. In addition, our approach demonstrates more desirable finite-sample performances over another existing two-stage estimation method proposed in Zhu H et al., (Commu Statistics-Theory Methods 51(22):7830-7845, 2021) . An R package PMLE4SCR is developed to implement our proposed method.

A LiDAR SLAM Algorithm Considering Dynamic Extraction of Feature Points in Underground Coal Mine

Abstract. To address the absence of GNSS signals in underground coal mines and the susceptibility of mainstream LiDAR SLAM to degradation due to insufficient feature constraints, this paper proposes a tightly-coupled SLAM algorithm incorporating LiDAR and IMU for use in such environments. The paper initially introduces a dynamic feature point extraction strategy, where the number of corner feature points can be dynamically adjusted based on the occurrence of degradation in the underground coal mine environment. This approach constructs a constraint matrix with rich and robust feature information, enhancing pose estimation accuracy in environments with inadequate feature constraints, mitigating degradation effects, and reducing global cumulative error. Subsequently, the consistent building of the underground coal mine environment is achieved through back-end factor graph optimization. Finally, to validate the effectiveness of the method, experimental analysis is conducted in an underground coal mine. The results demonstrate that LeGO_LOAM exhibits poor robustness in underground coal mines and fails to construct a globally consistent pose estimation and map. Conversely, the pose estimation error of the proposed method in this paper is 50.93% lower in the plane direction and 42.13% lower in the elevation direction compared to LIO_SAM. This underscores the method's significance as crucial technical support for the intelligent perception and positioning of robots in underground coal mines.

Psychometric Validation of the Purpose in Life Test-Short Form (PIL-SF) in Individuals Diagnosed with Severe Mental Illness.

Meaning in Life (MiL) represents a key variable in mental health models of personal recovery. There is a need for straightforward and concise instruments to assess this construct quantitatively in individuals diagnosed with severe mental illness (SMI). The aim of the present study was to test the psychometric properties of the Purpose in Life Test-Short Form (PIL-SF), a brief self-report measuring the presence of MiL, in a sample of individuals with SMI. The participants were 41 adults (21 women, 51.8% and 20 men, 48.2%) aged 18 to 65 years (M = 50.05; SD = 10.73) with a diagnosis of SMI (schizophrenia, 61%; bipolar disorder, 26.8%; borderline personality disorder, 7.3%; and major depression, 4.9%) and clinically stable. The PIL-SF, Satisfaction with Life Scale (SWLS), Oxford Happiness Questionnaire-6 Item (OHQ-6), Engagement in Meaningful Activities Survey (EMAS), and Seeking of Noetic Goals-8 Item (SONG-8) were used. Descriptive analysis, estimation of the internal consistency, and Confirmatory Factor Analysis of the PIL-SF were conducted. Furthermore, correlations between the PIL-SF, SWLS, OHQ-6, EMAS, and SONG-8 were calculated. The PIL-SF showed acceptable internal consistency (ω = 0.81) and an excellent fit as a unidimensional scale (CFI = 1.000, TLI = 1.070, RMSEA = 0.000, SRMR = 0.021), confirming its factorial structure. Regarding construct validity, correlations between the PIL-SF and the SWLS (ρ = 0.54, p < 0.001), the OHQ-6 (ρ = 0.52, p < 0.001), and EMAS (ρ = 0.44, p < 0.005) were positive and significant, whereas the correlations between the PIL-SF and the SONG-8NfM (ρ = -0.35, p < 0.025) were negative and significant. The Spanish version of the PIL-SF appears to be a reliable and valid instrument to measure the presence of MiL in adults with SMI.

Exploring the causal pathway from gut microbiota to polycystic ovary syndrome: A network Mendelian randomization study.

Polycystic ovary syndrome (PCOS) is a complicated endocrine and metabolic syndrome with unclear pathogenesis. The gut microbiota sheds light on the etiology and pathophysiology of PCOS. We used Mendelian randomization (MR) studies to systematically evaluate the pathological mechanism gut microbiota causally associated with PCOS risk. A network MR analysis was performed to estimate the causal effects of gut microbiota and risk factors on PCOS, as well as the mediation effect of risk factors linking gut microbiota to PCOS. The investigation of side effects for the important gut microbiota was subsequently broadened to include phenotypes by performing Phenowide-MR analysis for a range of diseases. Genus Sellimonas id.14369 were causally associated with reduced PCOS risk (odds ratio [OR] = 0.69, 95% confidence interval [CI]: 0.58-0.84, P = 1.22 × 10-4) after multiple testing correction. And Sellimonas retained consistent causal effect estimates after a series of sensitivity analyses. In addition, we observed an indirect effect of Sellimonas on PCOS through body mass index (BMI) using network MR (b = -0.05, 95% CI: -0.09 to -0.01), with a mediated proportion of 12.82% of the total effect. Further, Phenowide-MR analyses showed that the protective effects of Sellimonas on type 2 diabetes and depression (for type 2 diabetes: OR = 0.95, 95% CI: 0.90-0.99, P = .0366; for depression: OR = 0.99, 95% CI: 0.98-1.00, P = .0210). We summarized that the causal path between gut microbiota and type 2 diabetes are also jointly mediated by BMI. Sellimonas may be a protective factor of PCOS, which can affect the occurrence of PCOS through BMI, supporting future studies on the importance of addressing obesity and metabolic issues in preventing and managing PCOS.