Semiparametric Regression Model Research Articles

Complete f-moment convergence for arrays of random variables and its applications in semiparametric and EV regression models

Complete f-moment convergence for arrays of random variables and its applications in semiparametric and EV regression models

Semiparametric modeling for the cardiometabolic risk index and individual risk factors in the older adult population: A novel proposal.

Multivariate semiparametric regression models were used to study the association between the CMRI with the individual risk factors, which was achieved using secondary analysis the data from the SABE study (Survey on Health, Well-Being, and Aging in Colombia, 2015). The risk factors were selected through a stepwise procedure. The covariates included showed evidence of non-linear relationships with the CMRI, revealing non-linear interactions between: BMI and age (p< 0.00); arm and calf circumferences (p<0.00); age and females (p<0.00); walking speed and joint pain (p<0.02); and arm circumference and joint pain (p<0.00). Semiparametric modeling explained 24.5% of the observed deviance, which was higher than the 18.2% explained by the linear model.

Excess Mortality Risk Due to Heat Stress in Different Climatic Zones of India.

India is at a high risk of heat stress-induced health impacts and economic losses owing to its tropical climate, high population density, and inadequate adaptive planning. The health impacts of heat stress across climate zones in India have not been adequately explored. Here, we examine and report the vulnerability to heat stress in India using 42 years (1979-2020) of meteorological data from ERA-5 and developed climate-zone-specific percentile-based human comfort class thresholds. We found that the heat stress is usually 1-4 °C higher on heatwave (HW) days than on nonheatwave (NHW) days. However, the stress on NHW days remains considerable and cannot be neglected. We then showed the association of a newly formulated India heat index (IHI) with daily all-cause mortality in three cities - Delhi (semiarid), Varanasi (humid subtropical), and Chennai (tropical wet and dry), using a semiparametric quasi-Poisson regression model, adjusted for nonlinear confounding effects of time and PM2.5. The all-cause mortality risk was enhanced by 8.1% (95% confidence interval, CI: 6.0-10.3), 5.9% (4.6-7.2), and 8.0% (1.7-14.2) during "sweltering" days in Varanasi, Delhi, and Chennai, respectively, relative to "comfortable" days. Across four age groups, the impact was more severe in Varanasi (ranging from a 3.2 to 7.5% increase in mortality risk for a unit rise in IHI) than in Delhi (2.6-4.2% higher risk) and Chennai (0.9-5.7% higher risk). We observed a 3-6 days lag effect of heat stress on mortality in these cities. Our results reveal heterogeneity in heat stress impact across diverse climate zones in India and call for developing an early warning system keeping in mind these regional variations.

Pemodelan Kadar Hemoglobin pada Pasien Demam Berdarah di Kota Samarinda Menggunakan Regresi Semiparametrik Spline Truncated

This article discusses the innovation of statistical modeling in regression analysis with a semiparametric approach applied to health data. Regression analysis is a method in statistics that takes a lot of roles in statistical modeling. Regression analysis is used to model the relationship between the independent variable (x) and the dependent variable (y). There are three approaches to regression analysis, namely parametric, nonparametric, and a combination of the two, namely semiparametric. Semiparametric regression is used when the dependent variable has a known relationship with some of the independent variables and has an unknown pattern of a relationship with some of the other independent variables. The purpose of this study was to model hemoglobin levels in dengue fever patients, with the independent variables used being the number of hematocrits (x1) and the number of leukocytes (x2). The method used is spline truncated semiparametric regression. The truncated spline estimator was chosen for the nonparametric component because it has many advantages in modeling, one of which is being able to model patterns where the form of the relationship is unknown. The parameter estimation used is the maximum estimation. Selection of the optimal knot point using Generalized Cross-Validation (GCV). Based on the results of the analysis, the truncated spline semiparametric regression model was obtained which was applied to the hemoglobin level data in a model with three knots which have a coefficient of determination of 89.074%. Based on the results of testing the hypothesis simultaneously, it can be concluded that simultaneously the independent variable has a significant effect on the dependent variable. In the partial test, it is concluded that the variables x1 and x2 have a significant influence on the dependent variable y .

Complete f-moment convergence for m-asymptotic negatively associated random variables and related statistical applications

In this article, the complete f-moment convergence for m-asymptotic negatively associated random variables is investigated. As applications, we establish the strong consistency of the least square estimator in the simple linear errors-in-variables models and the complete consistency for estimator in the semiparametric regression model based on m-asymptotic negatively associated errors. We also give some simulations to assess the finite sample performance of the theoretical results.

Semi-parametric single-index predictive regression models with cointegrated regressors

This paper considers the estimation of a semi-parametric single-index regression model that allows for nonlinear predictive relationships. This model is useful for predicting financial asset returns, whose observed behaviour resembles a stationary process, if the multiple nonstationary predictors are cointegrated. The presence of cointegrated regressors imposes a single-index structure in the model, and this structure not only balances the nonstationarity properties of the multiple predictors with the stationarity properties of asset returns but also avoids the curse of dimensionality associated with nonparametric regression function estimation. An orthogonal series expansion is used to approximate the unknown link function for the single-index component. We consider the constrained nonlinear least squares estimator of the single-index (or the cointegrating) parameters and the plug-in estimator of the link function, and derive their asymptotic properties. In an empirical application, we find some evidence of in-sample nonlinear predictability of U.S. stock returns using cointegrated predictors. We also find that the single-index model in general produces better out-of-sample forecasts than both the historical average benchmark and the linear predictive regression model.

Complications and associated risk factors after surgical management of proximal femoral fractures.

This work aimed at answering the following research questions: 1) What is the rate of mechanical complications, nonunion and infection for head/neck femoral fractures, intertrochanteric fractures, and subtrochanteric fractures in the elderly USA population? and 2) Which factors influence adverse outcomes? Proximal femoral fractures occurred between 1 January 2009 and 31 December 2019 were identified from the Medicare Physician Service Records Data Base. The Kaplan-Meier method with Fine and Gray sub-distribution adaptation was used to determine rates for nonunion, infection, and mechanical complications. Semiparametric Cox regression model was applied incorporating 23 measures as covariates to identify risk factors. Union failure occured in 0.89% (95% confidence interval (CI) 0.83 to 0.95) after head/neck fracturs, in 0.92% (95% CI 0.84 to 1.01) after intertrochanteric fracture and in 1.99% (95% CI 1.69 to 2.33) after subtrochanteric fractures within 24 months. A fracture-related infection was more likely to occur after subtrochanteric fractures than after head/neck fractures (1.64% vs 1.59%, hazard ratio (HR) 1.01 (95% CI 0.87 to 1.17); p < 0.001) as well as after intertrochanteric fractures (1.64% vs 1.13%, HR 1.31 (95% CI 1.12 to 1.52); p < 0.001). Anticoagulant use, cerebrovascular disease, a concomitant fracture, diabetes mellitus, hypertension, obesity, open fracture, and rheumatoid disease was identified as risk factors. Mechanical complications after 24 months were most common after head/neck fractures with 3.52% (95% CI 3.41 to 3.64; currently at risk: 48,282). The determination of complication rates for each fracture type can be useful for informed patient-clinician communication. Risk factors for complications could be identified for distinct proximal femur fractures in elderly patients, which are accessible for therapeutical treatment in the management.

Exploring the sustainable growth pathway of wind power in China: Using the semiparametric regression model

China not only faces severe energy shortages, but also enormous pressure to reduce CO2 emissions. The objective of this article is to explore effective pathways and policies to promote wind power development in China. The interesting results include: (1) the impact of environmental regulations and green finance on wind power is nonlinear. Specifically, environmental regulations exert an inverted U-shaped impact in the central and eastern regions, and a U-shaped effect in the western region. (2) Green finance produces a U-shaped impact on wind power in the eastern region, and an N-shaped effect in the central and western regions. The impact of other economic factors shows a linear pattern. (3) Urbanization has the greatest impact on wind power in the central region, because of greater investment in technological talents and faster growth in high-tech industries. (4) Green technology innovation has made the greatest contribution to wind power in the western region, because of the rapidly growing green energy technology. (5) Financial decentralization has the smaller driving effect on wind power development in the eastern region, due to small financial investment. (6) Wind power prices have a driving effect on wind power development in the western region, but not played a driving role in the central and eastern regions. Finally, this article proposes promoting policies based on estimated results.

Local Gaussian Process Extrapolation for BART Models with Applications to Causal Inference

Bayesian additive regression trees (BART) is a semi-parametric regression model offering state-of-the-art performance on out-of-sample prediction. Despite this success, standard implementations of BART typically suffer from inaccurate prediction and overly narrow prediction intervals at points outside the range of the training data. This article proposes a novel extrapolation strategy that grafts Gaussian processes to the leaf nodes in BART for predicting points outside the range of the observed data. The new method is compared to standard BART implementations and recent frequentist resampling-based methods for predictive inference. We apply the new approach to a challenging problem from causal inference, wherein for some regions of predictor space, only treated or untreated units are observed (but not both). In simulation studies, the new approach boasts superior performance compared to popular alternatives, such as Jackknife+. Supplementary materials for this article are available online.

Multivariate Bayesian Semiparametric Regression Model for Forecasting and Mapping HIV and TB Risks in West Java, Indonesia

Multivariate “Bayesian” regression via a shared component model has gained popularity in recent years, particularly in modeling and mapping the risks associated with multiple diseases. This method integrates joint outcomes, fixed effects of covariates, and random effects involving spatial and temporal components and their interactions. A shared spatial–temporal component considers correlations between the joint outcomes. Notably, due to spatial–temporal variations, certain covariates may exhibit nonlinear effects, necessitating the use of semiparametric regression models. Sometimes, choropleth maps based on regional data that is aggregated by administrative regions do not adequately depict infectious disease transmission. To counteract this, we combine the area-to-point geostatistical model with inverse distance weighted (IDW) interpolation for high-resolution mapping based on areal data. Additionally, to develop an effective and efficient early warning system for controlling disease transmission, it is crucial to forecast disease risk for a future time. Our study focuses on developing a novel multivariate Bayesian semiparametric regression model for forecasting and mapping HIV and TB risk in West Java, Indonesia, at fine-scale resolution. This novel approach combines multivariate Bayesian semiparametric regression with geostatistical interpolation, utilizing population density and the Human Development Index (HDI) as risk factors. According to an examination of annual data from 2017 to 2021, HIV and TB consistently exhibit recognizable spatial patterns, validating the suitability of multivariate modeling. The multivariate Bayesian semiparametric model indicates significant linear effects of higher population density on elevating HIV and TB risks, whereas the impact of the HDI varies over time and space. Mapping of HIV and TB risks in 2022 using isopleth maps shows a clear HIV and TB transmission pattern in West Java, Indonesia.

Impact of Industrialization on Household Income in Rural China, 1988-2018

&lt;p style="text-align: justify;"&gt;This paper uses China Household Income Project rural household survey data spanning 30 years (1988-2018) to study the impact of China&amp;rsquo;s industrialization on rural household income. In the semiparametric regression model, we use the rural households&amp;rsquo; distance to the manufacturing hub to analyze the spillover effect of industrialization on their income and the spatial attenuation. The results show that the income gap between rural households in Guangdong and other provinces increased from 1988 to 2002, but continuously decreased from 2002 to 2018; the rural households&amp;rsquo; distance to the Pearl River Delta or the Yangtze River Delta, the two international manufacturing hubs, has a significant impact on their income, which has been increasing from 1988 to 1995, but has been decreasing since the new century. On the contrary, the influence of the rural households&amp;rsquo; distance to the provincial capital on their income has been increasing since the new century.&lt;/p&gt;

Association of dietary mineral mixture with depressive symptoms: A combination of Bayesian approaches

The relationships between mixtures of multiple minerals and depression have not been explored. Therefore, we analyzed the relationship between the mixture of nine dietary minerals [calcium (Ca), phosphorus, magnesium (Mg), iron (Fe), zinc, copper (Cu), sodium, potassium (K), and selenium (Se)] and depressive symptoms in the general population. We screened 20,342 participants from the National Health and Nutrition Examination Survey (NHANES) 2007–2018. We fitted the general linear regression, Bayesian kernel machine regression (BKMR), and Bayesian semiparametric regression models to explore associations and interactions. We obtained the relative importance of dietary minerals by calculating posterior inclusion probabilities (PIPs). The dietary intakes of minerals were obtained using the 24-h dietary recall interview, and depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9). The linear analysis showed that nine minerals were negatively associated with PHQ-9 scores. The BKMR analysis showed a negative association between the dietary mineral mixture and PHQ-9 scores, with Se having the largest PIP at 1.0000, followed by K (0.7784). We also observed potential interactions between Ca and Fe, Se and Fe, and K and Mg. Among them, the interaction of Ca and Fe had the largest PIP of 0.986. In addition, the overall effect was more pronounced in females than males, and Cu's PIP (0.8376) was higher in females. Two sensitivity analyses showed that our results were robust. Our study provides a basis for formulating nutritional intervention programs for depression in the future.

Instrumental variable quantile regression under random right censoring

Summary This paper studies a semiparametric quantile regression model with endogenous variables and random right censoring. The endogeneity issue is solved using instrumental variables. It is assumed that the structural quantile of the logarithm of the outcome variable is linear in the covariates and censoring is independent. The regressors and instruments can be either continuous or discrete. The specification generates a continuum of equations of which the quantile regression coefficients are a solution. Identification is obtained when this system of equations has a unique solution. Our estimation procedure solves an empirical analogue of the system of equations. We derive conditions under which the estimator is asymptotically normal and prove the validity of a bootstrap procedure for inference. The finite sample performance of the approach is evaluated through numerical simulations. An application to the national Job Training Partnership Act study illustrates the method.

Complete f-moment convergence for weighted sums of asymptotically almost negatively associated random variables and its application in semiparametric regression models

In this paper, we investigate the complete f-moment convergence for weighted sums of asymptotically almost negatively associated (AANA, for short) random variables. Our results improve and generalize the corresponding ones of [M.M. Xi, X. Deng, X.J. Wang, and Z.Y. Cheng, convergence and complete convergence for weighted sums of AANA random variables, Commun. Stat. Theory Methods 47(22) (2018), pp. 5604–5613]. As an application of our main results, some results on the complete consistency for the estimator in semiparametric regression models are obtained and a simulation study is provided to assess the finite sample performance of the theoretical results.

Potts-Cox survival regression

A Bayesian semi-parametric survival regression model with latent partitions is introduced. Its goal is to predict survival and to cluster survival patients within the context of building prognosis systems. In order to drive cluster formation on individuals, the Potts random partition model is chosen as a prior on the covariates space. For any given partition, the proposed model assumes an interval-wise exponential distribution for the baseline hazard rate. The number of intervals is unknown. It can be estimated with a fused-lasso type penalty given by a sequential double exponential prior. Estimation and inference are done with the aid of Markov chain Monte Carlo. To simplify the computations, the Laplace's integral approximation method is used to estimate some constants and to propose parameter updates within Markov chain Monte Carlo. The methodology is illustrated with an application to cancer survival.

Changing the Discount Rate by Adjusting the Pure Rate of Time Preference

The Ramsey (1928) equation decomposes the real discount rate into the pure rate of time preference plus a term that accounts for the changing marginal utility of consumption. Discussions about the appropriate discount rate to apply in Cost Benefit Analysis sometimes refer to variations induced by alternative values of the pure rate of time preference as if the two vary on a one-to-one basis. But the optimal consumption path, which determines the marginal product of capital and hence the discount rate, depends on the rate of time preference. Hence the discount rate depends on time preference through the marginal utility term. We derive an analytical expression of this relationship and show that the derivative of the discount rate with respect to time preference only equals unity in the steady state and converges from below. We estimate the derivative using US data from 1930 to 2015. Based on a semi-parametric regression model with time-varying coefficients we find it is about 0.9, but we cannot rule out 1.0 being included in the 95% confidence interval. The implied pure rate of time preference after 1980 is about 1.6 percent.

P-752 Cervical cerclage versus vaginal progesterone in the prevention of preterm birth in singleton pregnancies among women with a short cervical length: A retrospective cohort study

Abstract Study question To compare the maternal and neonatal effects of cervical cerclage and vaginal progesterone in preventing singleton preterm births in women with a short cervical length. Summary answer Cervical cerclage showed no benefit in preventing preterm birth; however, it prolonged gestation by 39 days compared to vaginal progesterone treatment. What is known already Clinically, there may be a bias toward the choice of cervical cerclage if the cervix is short (&amp;lt;10 mm) and, similarly, toward vaginal progesterone use if the cervix is long (&amp;gt;30 mm). However, there are no clear treatment guidelines for the best options for singleton pregnant women with a CL between 10 and 30 mm. Furthermore, there is insufficient evidence to determine whether one treatment is superior to the other in preventing PTB and decreasing adverse neonatal outcomes in singleton pregnancies in such cases. Study design, size, duration Asymptomatic pregnant women having singleton pregnancies, with a CL between 10 and 30 mm, measured using transvaginal ultrasound at 12–26 weeks of gestation, who delivered at our hospital between January 2009 and August 2020, were included in our study. Based on the inclusion and exclusion criteria, 267 singleton pregnant women, of whom 116 were treated with cervical cerclage and 151 with vaginal progesterone, were finally included. Participants/materials, setting, methods Independent sample t-tests and differential analysis were used to analyze the data based on their distribution. The Cox semi-parametric regression model was used to analyze the gestation weeks at delivery and neonatal outcomes. The decision regarding cesarean delivery was based on a binary logistic regression model. Main results and the role of chance In the initial analysis, the number of preterm births was significantly higher in the cerclage group than in the vaginal progesterone group; however, after multivariate adjustment for confounding factors, the latency period from diagnosis to delivery was found to be significantly prolonged in the cerclage group. Limitations, reasons for caution There are currently no uniform criteria for determining whether study subjects are eligible for cervical cerclage or vaginal progesterone treatment, which may present some cases of noncompliance. Wider implications of the findings In our study, we found that cervical cerclage did not have significant advantages over vaginal progesterone treatment in preventing PTB. However, upon adjustment for confounding factors, it was shown to prolong the gestational age by 39 days compared to vaginal progesterone treatment. Trial registration number non-clinical trials

Endogeneity in semiparametric threshold regression models with two threshold variables

ABSTRACT This article considers a semiparametric threshold regression model with two threshold variables. The proposed model allows endogenous threshold variables and endogenous slope regressors. Under the diminishing threshold effects framework, we derive consistency and asymptotic results of our proposed estimator for weakly dependent data. We study the finite sample performance of our proposed estimator via small Monte Carlo simulations and apply our model to classify economic growth regimes based on both national public debt and national external debt.

Nonparametric tests for semiparametric regression models

Semiparametric regression models have received considerable attention over the last decades, because of their flexibility and their good finite sample performances. Here we propose an innovative nonparametric test for the linear part of the models, based on random sign-flipping of an appropriate transformation of the residuals, that exploits a spectral decomposition of the residualizing matrix associated with the nonparametric part of the model. The test can be applied to a vast class of extensively used semiparametric regression models with roughness penalties, with nonparametric components defined over one-dimensional, as well as over multi-dimensional domains, including, for instance, models based on univariate or multivariate splines. We prove the good asymptotic properties of the proposed test. Moreover, by means of extensive simulation studies, we show the superiority of the proposed test with respect to current parametric alternatives, demonstrating its excellent control of the Type I error, accompanied by a good power, even in challenging data scenarios, where instead current parametric alternatives fail.

Survival of the fittest: do firms actively or passively learn survival?

PurposeIn the corporate world, there is no certainty of survival. This research aims to identify firm-level factors that increase or decrease a firm's probability of exit and survival.Design/methodology/approachThe study examines 153 listed textile sector firms in Pakistan over a 10-year period from 2009 to 2018, comprising 1,413 observations. The semi-parametric Cox regression model is used to process the results.FindingsThe study finds that larger and exporting firms are more likely to survive, while those with a high ratio of fixed assets to total assets, high expenditure on advertising and variable costs are less likely to survive. The relationship between age and firm survival is inconclusive.Research limitations/implicationsAdaptability to the external environment provides a competitive advantage that is crucial for textile firms to reduce their chances of exit. The research is valuable for strategic managers and policymakers to identify focus areas to prevent firm exit.Originality/valueThis study supports the active learning theory, which suggests that new entrants in the textile sector of Pakistan should focus on becoming active market players, increasing efficiency and reducing variable costs to survive.