Semiparametric Regression Research Articles

Model-Free Radio Map Estimation in Massive MIMO Systems via Semi-Parametric Gaussian Regression

Accurate radio maps will be very much needed to provide environmental awareness and effectively manage future wireless networks. Most of the research so far has focused on developing power mapping algorithms for single and omnidirectional antenna systems. In this letter, we investigate the construction of crowdsourcing-based radio maps for 5G cellular systems with massive directional antenna arrays (spatial multiplexing), proposing an original technique based on semi-parametric Gaussian regression. The proposed method is <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">model-free</i> and provides highly accurate estimates of the radio maps, outperforming fully parametric and non-parametric solutions.

A preliminary investigation to determine the suitability of pigs as human analogues for post-mortem lipid analysis

The determination of time since death is a major challenge to law enforcement when faced with the discovery of human remains. This is due to the fact that decomposition is a complex, dynamic process influenced by several abiotic and biotic factors. For decades, post-mortem decomposition studies have used pigs as human analogues due to ethical and legal restrictions surrounding the use of human cadavers for such research. However, few comparative studies have been conducted to assess the suitability of these analogues. Recent forensic studies have successfully demonstrated the use of post-mortem lipids in textiles as a method to obtain vital information about decomposition process. The current investigation involved two studies: Trial 1 (summer) and Trial 2 (winter). Each trial with n = 1 human cadaver and n = 2 pigs. Samples were collected over a timeline of 105 days post-placement and analyzed using attenuated total reflectance (ATR-) Fourier-transform infrared (FTIR) spectroscopy. The data was then statistically assessed using functional principal component analysis, semi-parametric regression modeling and analysis of variance. The results demonstrated a clear statistically significant interspecies difference between pigs and humans in both trials. The preliminary implications of this study suggest that pigs are not suitable analogues for humans in decomposition research and have broader implications that caution the direct translation of decomposition data obtained from pigs to real human casework, particularly with respect to time since death estimations.

Estimation of conditional cumulative incidence functions under generalized semiparametric regression models with missing covariates, with application to analysis of biomarker correlates in vaccine trials.

This article studies generalized semiparametric regression models for conditional cumulative incidence functions with competing risks data when covariates are missing by sampling design or happenstance. A doubly-robust augmented inverse probability weighted complete-case (AIPW) approach to estimation and inference is investigated. This approach modifies IPW complete-case estimating equations by exploiting the key features in the relationship between the missing covariates and the phase-one data to improve efficiency. An iterative numerical procedure is derived to solve the nonlinear estimating equations. The asymptotic properties of the proposed estimators are established. A simulation study examining the finite-sample performances of the proposed estimators shows that the AIPW estimators are more efficient than the IPW estimators. The developed method is applied to the RV144 HIV-1 vaccine efficacy trial to investigate vaccine-induced IgG binding antibodies to HIV-1 as correlates of acquisition of HIV-1 infection while taking account of whether the HIV-1 sequences are near or far from the HIV-1 sequences represented in the vaccine construct.

How do women on corporate boards shape corporate social performance? Evidence drawn from semiparametric regression

This study re-examines the relationship between women on corporate boards (WOCB) and corporate social performance (CSP) for a sample of companies from the Fortune 1000 ranking over the period 2004 to 2018 (ranked from 501 to 1000). To take into account the complex and non-linear relationship as well as endogeneity issues, we use a two-stage generalised additive model. This contribution is significant, as many authors have demonstrated the non-linearity of factors influencing performance, whether of a financial or social nature. Consistent with token and critical mass theories, our results shows that the effects of WOCB on CSP vary significantly depending on their numerical representation and that there are departures from linearity. Our findings provide explanations for the existing mixed empirical results, which all rely on parametric methods. As such, we suggest the use of semiparametric methods takin into account endogeneity issues to assess WOCB–CSP relationship. This study sheds some new light on that relationship, which remains a controversial issue.

Determinants of Stroke Mortality through Survival Models: The Case of Mettu Karl Referral Hospital, Mettu, Ethiopia.

Introduction Every year worldwide, between five to six million deaths are associated with stroke; on average, one stroke-related death occurs every four minutes. In Ethiopia, stroke is a frequent cause of mortality and morbidity from noncommunicable diseases. Therefore, this study was aimed at determining factors associated to stroke mortality through survival models in Mettu Karl Referral Hospital. Methods This study was conducted from September 1, 2014, to April 1, 2017, and encompassed 202 stroke patients at Mettu Karl Referral Hospital. The Cox semiparametric regression was used for analyzing survival analysis of stroke patients using R software. Results A total of 202 stroke patients were included in the study, and among those patients, 72.8% and 27.2% were censored and died, respectively. According to the result of Cox semiparametric regression model, sex of patients, hypertension, baseline complication, and stroke type had significant effect on survival of the stroke patient at 5% significance level. Conclusion The results from Cox semiparametric regression model indicated that sex of patients, hypertension, baseline complication, and stroke type were major factors related to the survival time of stroke patients. The researcher recommends that the people should be aware on the burden of those risk factors and well informed about the disease.

Spatiotemporal clustering and correlates of childhood stunting in Ghana: Analysis of the fixed and nonlinear associative effects of socio-demographic and socio-ecological factors.

Childhood stunting remains a major public health issue in many low- and middle-income countries. In Ghana, the progress made is insufficient to reach the targets set by the 2025 World Health Assembly and the 2030 United Nations Sustainable Development Goals. Although studies have examined the socio-demographic determinants of childhood stunting, there has not been any systematic study to examine the spatial associative effects of the socio-demographic and socio-ecological factors at the district level, where health programmes are implemented and monitored. Bayesian geo-additive semiparametric regression technique was used to analyse five conservative rounds of Demographic and Health Surveys in Ghana, with socio-ecological covariates derived from the Demographic and Health Survey Program Geospatial Covariate datasets to examine the temporal trends in childhood stunting, the extent of geospatial clustering at the district level and their associative relationships with socio-demographic and socio-ecological factors. The findings show that childhood stunting in Ghana is not spatially randomly distributed but clustered. Clustering of high childhood stunting was observed amongst districts in the Upper West, Upper East, Northern, North East, Savannah, and Western North regions, whilst clustering of low childhood stunting was observed in districts in the Greater Accra, Volta, Bono and the Eastern regions. Whist socio-demographic factors were predominantly associated with clustering of districts with high childhood stunting, the socio-ecological factors were mainly associated with clustering of districts with low childhood stunting. The socio-ecological factors identified to have a nonlinear associative effect with childhood stunting were Insecticide Treated Net (ITN) coverage, nightlight composite, travel time to a main settlement and population density. The findings suggest that targeted interventions at the district level are essential to reducing childhood stunting in Ghana.

Consistency and Asymptotic Normality of Estimator for Parameters in Multiresponse Multipredictor Semiparametric Regression Model

A multiresponse multipredictor semiparametric regression (MMSR) model is a combination of parametric and nonparametric regressions models with more than one predictor and response variables where there is correlation between responses. Due to this correlation we need to construct a symmetric weight matrix. This is one of the things that distinguishes it from the classical method, which uses a parametric regression approach. In this study, we theoretically developed a method of determining a confidence interval for parameters in a MMSR model based on a truncated spline, and investigating asymptotic properties of estimator for parameters in a MMSR model, especially consistency and asymptotic normality. The weighted least squares method was used to estimate the MMSR model. Next, we applied a pivotal quantity method, a Cramer–Wold theorem, and a Slutsky theorem to determine the confidence interval, investigate consistency, and asymptotic normality properties of estimator for parameters in a MMSR model. The obtained results were that the estimated regression function is linear to observation. We also obtained a 1001−α% confidence interval for parameters in the MMSR model, and the estimator for parameters in MMSR model was consistent and asymptotically normally distributed. In the future, these obtained results can be used as a theoretical basis in designing a standard toddlers growth chart to assess nutritional status.

Mixture regression for longitudinal data based on joint mean–covariance model

In the process of modeling longitudinal data, we focus on the case that the studied population is comprised of different groups of individuals and individuals within the same group share the similar kind of mean progression trajectories, where finite mixture models (FMM) are often used to address this kind of unobserved heterogeneity in terms of mean. Existing methods, such as parametric and semiparametric mixture regression, usually model the mean in each subpopulation with assumption that observations sharing a common trajectory are independent or their covariance structure is pre-specified, but less research considers modeling of covariance structures while accounting for heterogeneity. In this paper, we introduce a joint model which models the mean and covariance structures simultaneously in a finite normal mixture regression, demonstrating how important the within-subject correlation is in clustering longitudinal data. Model parameters are estimated with an iteratively re-weighted least squares EM (IRLS-EM) algorithm. Our estimators are shown to be consistent and asymptotically normal. We can identify different mean trajectories and covariance structures in all clusters. Simulations show that the proposed method performs well and gives more accurate clustering results by introducing covariance modeling. Real data analysis is also used to illustrate the usefulness of the proposed method, and it presents good performance in clustering COVID-19 deaths for European countries in terms of progression trajectory.

Multivariate Functional Kernel Machine Regression and Sparse Functional Feature Selection.

Motivated by mobile devices that record data at a high frequency, we propose a new methodological framework for analyzing a semi-parametric regression model that allow us to study a nonlinear relationship between a scalar response and multiple functional predictors in the presence of scalar covariates. Utilizing functional principal component analysis (FPCA) and the least-squares kernel machine method (LSKM), we are able to substantially extend the framework of semi-parametric regression models of scalar responses on scalar predictors by allowing multiple functional predictors to enter the nonlinear model. Regularization is established for feature selection in the setting of reproducing kernel Hilbert spaces. Our method performs simultaneously model fitting and variable selection on functional features. For the implementation, we propose an effective algorithm to solve related optimization problems in that iterations take place between both linear mixed-effects models and a variable selection method (e.g., sparse group lasso). We show algorithmic convergence results and theoretical guarantees for the proposed methodology. We illustrate its performance through simulation experiments and an analysis of accelerometer data.

Analysis of composite endpoints with component-wise censoring in the presence of differential visit schedules.

Composite endpoints are very common in clinical research, such as recurrence-free survival in oncology research, defined as the earliest of either death or disease recurrence. Because of the way data are collected in such studies, component-wise censoring is common, where, for example, recurrence is an interval-censored event and death is a right-censored event. However, a common way to analyze such component-wise censored composite endpoints is to treat them as right-censored, with the date at which the non-fatal event was detected serving as the date the event occurred. This approach is known to introduce upward bias when the Kaplan-Meier estimator is applied, but has more complex impact on semi-parametric regression approaches. In this article we compare the performance of the Cox model estimators for right-censored data and the Cox model estimators for interval-censored data in the context of component-wise censored data where the visit process differs across levels of a covariate of interest, a common scenario in observational data. We additionally examine estimators of the cause-specific hazard when applied to the individual components of such component-wise censored composite endpoints. We found that when visit schedules differed according to levels of a covariate of interest, the Cox model estimators for right-censored data and the estimators for cause-specific hazards were increasingly biased as the frequency of visits decreased. The Cox model estimator for interval-censored data with censoring at the last disease-free date is recommended for use in the presence of differential visit schedules.

Competing risks analysis with missing cause-of-failure-penalized likelihood estimation of cause-specific Cox models.

Competing risks models are attractive tools to analyze time-to-event data where several causes of an event are competing. However, a complexity may arise when, for instance, some subjects experience the event of interest but the causes are not known. Assuming that unknown causes of events are missing at random, we developed a novel constrained maximum penalized likelihood method for fitting semi-parametric cause-specific Cox regression models. Here, penalty functions were used to smooth the baseline hazards. An appealing feature of this approach is that all the relevant estimands in competing risks models are estimated including cause-specific hazard ratios, cause-specific baseline hazards, and cumulative incidence functions. Asymptotic results for these estimators were also developed, allowing for direct inferences. The proposed method was compared with some existing methods through a simulation study. A real data example was analyzed using the new method to evaluate the association of age at diagnosis with melanoma-death and non-melanoma-death in patients diagnosed with thin melanoma (tumour thickness 1.0 mm). An R function for our proposed method is currently available on GitHub and will be included in the R package "survivalMPL" at CRAN.

R&D expenditure and managerial ownership: evidence from firms of high-vs-low R&D intensity

PurposeCorporate managers spend on research and development (R&D) for reasons of growth and survival. However, they may be less willing to invest in R&D because of its long-term horizon, high failure rate and uncertain outcomes. This study aims to explore the extent to which managerial ownership influences R&D expenditure decisions.Design/methodology/approachApart from the linear regression models, this study uses a semi-parametric quantile regression analysis for a sample of German non-financial firms throughout 2009–2018.FindingsThis study finds a nonmonotonic sensitivity of R&D spending to the level of managerial ownership over various quantiles of R&D distribution. That is, managerial ownership increases the expenditure on R&D at low R&D intensity firms. However, it decreases the expenditure on R&D at high R&D intensity firms. These results suggest the presence of a maximum level of R&D expenditure, after which owner-managers would be unwilling to spend on R&D.Practical implicationsThe results confirm the importance of corporate ownership structure for firm R&D and innovation activities. It provides an implication for corporate policymakers to reform the corporate ownership structures to encourage corporate managers and owners to invest in R&D projects.Originality/valueThis study offers two distinct contributions study. First, it provides the first German shred of evidence on the nonlinear relationship between managerial ownership and R&D expenditure decisions by distinguishing between high and low R&D intensity firms. Second, unlike prior research, it uses a semi-parametric quantile regression analysis. This method is more efficient than least-squares estimators and produces robust estimators to heteroscedasticity of the residuals.

Proposed Models for Comprehensive Automobile Insurance Ratemaking in Egypt with Parametric and Semi-Parametric Regression: A case study

Proposed Models for Comprehensive Automobile Insurance Ratemaking in Egypt with Parametric and Semi-Parametric Regression: A case study

Semiparametric Distribution Regression with Instruments and Monotonicity

Semiparametric Distribution Regression with Instruments and Monotonicity

The ensemble conditional variance estimator for sufficient dimension reduction

Ensemble Conditional Variance Estimation (ECVE) is a novel sufficient dimension reduction (SDR) method in regressions with continuous response and predictors. ECVE applies to general non-additive error regression models and operates under the assumption that the predictors can be replaced by a lower dimensional projection without loss of information. It is a semiparametric forward regression model-based exhaustive sufficient dimension reduction estimation method that is shown to be consistent under mild assumptions. ECVE outperforms central subspace mean average variance estimation (csMAVE), its main competitor, under several simulation settings and in a benchmark data set analysis.

Prognosis and Overall Survival of Thalassemia Patients using Semi-Parametric Mixture Cure-Rate Model: a 30-Year Retrospective Cohort Study

Understanding the determinants of long-term overall survival (OS) of thalassemia patients (TPs) is the mainstay of care. As a retrospective survey, we assessed the data of 769 TPs who had regular follow-up and blood transfusion for at least 30 years from 1990 - 2019. We utilized semi-parametric proportional hazards mixture cure-rate regression to discover the factors with a significant effect on short- and long-term OS separately. The 25- and 30-year OS for the TPs were calculated to be 98.7% and 90.4%, respectively. Each five-year age escalation was associated with a 30% decrease in the probability of being short-term survivors (HR = 1.06, p = 0.047). Parental family relationship influenced both cured (OR = 3.00, p = 0.017) and uncured (HR = 0.50, p = 0.046) TPs. Moreover, the type of iron chelation drug, liver iron concentration, and normal EF results had a significant effect on long-term OS. Aging, parental consanguinity, liver and cardiac siderosis, higher ferritin levels, and low hemoglobin level were associated with poorer prognosis in TPs. However, deferoxamine followed by multiple drugs as iron chelation, severe liver siderosis, and abnormal EF declined the probability of long-term OS among TPs. This can be considered by health policy decision-makers to enforce the screening program more strictly.

Semi-parametric regression function estimation for environmental pollution with measurement error using artificial flower pollination algorithm

Artificial Intelligence Algorithms have been used in recent years in many scientific fields. We suggest employing flower pollination algorithm in the environmental field to find the best estimate of the semi-parametric regression function with measurement errors in the explanatory variables and the dependent variable, where measurement errors appear frequently in fields such as chemistry, biological sciences, medicine, and epidemiological studies, rather than an exact measurement. We estimate the regression function of the semi-parametric model by estimating the parametric model and estimating the non-parametric model, the parametric model is estimated by using an instrumental variables method (Wald method, Bartlett's method, and Durbin's method), The non-parametric model is estimated by using kernel smoothing (Nadaraya Watson), K-Nearest Neighbor smoothing and Median smoothing. The Flower Pollination algorithms were employed and structured in building the ecological model and estimating the semi-parametric regression function with measurement errors in the explanatory and dependent variables, then compare the models to choose the best model used in the environmental scope measurement errors, where the comparison between the models is done using the mean square error (MSE). These methods were applied to real data on environmental pollution/ air pollution in the city of Baghdad, and the most important conclusions that we reached when using statistical methods in estimating parameters and choosing the best model, we found that the Median-Durbin model is the best as it has less MSE, but when using flower The pollination algorithm showed that the Median-Wald model is the best because it has the lowest MSE, and when we compare the statistical methods with the FPA in selecting semi-parametric models, we notice the superiority of the FP algorithm in all methods and for all models.

Confidence interval of parameters in multiresponse multipredictor semiparametric regression model for longitudinal data based on truncated spline estimator

In this paper, we provide a theoretical discussion on estimating confidence interval of parameters in a multiresponse multipredictor semiparametric regression (MMSR) model for longitudinal data. The MMSR model consists of two components namely a parametric component and a nonparametric component. In consequently, estimating the MMSR model is equivalent to estimating the parametric and nonparametric components. Estimating the parametric component is equivalent to estimating parameters of the model, while estimating the nonparametric component is estimating unknown smooth function. In this paper, we estimate the parametric and nonparametric components using a weighted least square method and a smoothing technique namely truncated spline, respectively. Next, we estimate the confidence interval of parameters in the MMSR model using pivotal quantity and Lagrange multiplier functions. The results of this study can be applied to the Covid-19 data that is to model the case growth rate (CGR) and case fatality rate (CFR) of Covid-19 which are influenced by many variables including comorbid, age, gender, temperature, self-isolation, isolation in hospital, and others.

Semiparametric Regression Analysis of Mixed Recurrent Event and Panel Count Data with Multiple Causes of Failure

Semiparametric Regression Analysis of Mixed Recurrent Event and Panel Count Data with Multiple Causes of Failure

Robust ridge estimator in censored semiparametric linear models

Besides censoring, multicollinearity and outliers are two common problem in regression analysis. In this paper we propose a family of robust ridge estimators for the censored semiparametric regression models. The proposed robust estimators is based on least trimmed squares (LTS) method. This method is insensitive to corruption due to outliers, provided that the outliers constitute less than of the set, in other words, LTS is a robust estimator with a breakdown point. The FAST-LTS algorithm is developed for the computation of the estimators. Furthermore, a robust method for the estimate of shrinkage parameters is suggested. Monté-Carlo simulation study demonstrates the merit of the new method in the aspect of solving the multicollinearity and sensitivity to outliers over the ordinary least squares estimation. Finally, an example of real data is given for illustration.