Number Of Explanatory Variables Research Articles

A high‐dimensional multinomial logit model

SummaryThe number of parameters in a standard multinomial logit model increases linearly with the number of choice alternatives and number of explanatory variables. Because many modern applications involve large choice sets with categorical explanatory variables, which enter the model as large sets of binary dummies, the number of parameters in a multinomial logit model is often large. This paper proposes a new method for data‐driven two‐way parameter clustering over outcome categories and explanatory dummy categories in a multinomial logit model. A Bayesian Dirichlet process mixture model encourages parameters to cluster over the categories, which reduces the number of unique model parameters and provides interpretable clusters of categories. In an empirical application, we estimate the holiday preferences of 11 household types over 49 holiday destinations and identify a small number of household segments with different preferences across clusters of holiday destinations.

Multi-Scale Effects of Landscape Pattern on Soundscape Perception in Residential Green Spaces

Soundscape quality in green spaces of residential areas directly contributes to residents’ quality of life. It has close relationships with landscape characteristics, which should be considered in landscape planning and design processes in residential areas. Accordingly, this study proposed a new perspective on the interrelationships between soundscape perception and landscape pattern on multi-scale, based on a case study of 30 residential green spaces in Fuzhou, China. Percentage of Landscape (PLAND), Patch Density (PD), Landscape Shape Index (LSI), and Patch cohesion index (COHESION) were utilized to represent the landscape pattern of vegetation, buildings, and roads in the residential areas. Soundscape perception was interpreted using the sound dominant degree (SDD) of sound sources and overall soundscape quality. The examined spatial scales range from 20 m to 180 m, with concentric circles spaced 20 m apart for each sampling point. Correlation analyses indicated that most landscape indices of vegetation and buildings were correlated with these soundscape perception indicators, while limited landscape indices of roads were associated with them. Based on the multi-scale landscape indices, multiple linear regression models for the SDD of sound sources and overall soundscape quality were established, confirming that the scale effect of landscape patterns can affect soundscape perception. Expressly, results indicated that these models were chiefly influenced by the landscape indices at a scale less than 120 m, but the scale effect of landscape pattern on the SDD of birdsong, pleasantness, and quietness was not so evident. Furthermore, we found that the number of explanatory variables may somewhat affect the model performance. The overall interpretability of these landscape indices for the SDD of sound sources was better than that of overall soundscape quality, implying the complexity of the latter. This study offers a fresh insight into the relationship between landscapes and soundscapes at varying scales. The findings can provide useful information for the promotion strategies of landscapes and soundscapes, especially in residential green spaces.

Generalized Support Vector Regression and Symmetry Functional Regression Approaches to Model the High-Dimensional Data

The analysis of the high-dimensional dataset when the number of explanatory variables is greater than the observations using classical regression approaches is not applicable and the results may be misleading. In this research, we proposed to analyze such data by introducing modern and up-to-date techniques such as support vector regression, symmetry functional regression, ridge, and lasso regression methods. In this study, we developed the support vector regression approach called generalized support vector regression to provide more efficient shrinkage estimation and variable selection in high-dimensional datasets. The generalized support vector regression can improve the performance of the support vector regression by employing an accurate algorithm for obtaining the optimum value of the penalty parameter using a cross-validation score, which is an asymptotically unbiased feasible estimator of the risk function. In this regard, using the proposed methods to analyze two real high-dimensional datasets (yeast gene data and riboflavin data) and a simulated dataset, the most efficient model is determined based on three criteria (correlation squared, mean squared error, and mean absolute error percentage deviation) according to the type of datasets. On the basis of the above criteria, the efficiency of the proposed estimators is evaluated.

Kernel semi-parametric model improvement based on quasi-oppositional learning pelican optimization algorithm

Statistical modeling is essential in many scientific research areas because it explains the relationship between the response variable of interest and a number of explanatory variables. However, it is not easy to determine the optimal model beforehand. Therefore, in this paper, we look at how to choose a hyper-parameter in a kernel semi-parametric regression model. A quasi-oppositional learning pelican optimization algorithm strategy is used to select the smoothness parameter. In comparison to other competitor approaches, simulation results revealed that the suggested method, the quasi-oppositional learning pelican optimization algorithm, is superior in terms of MSE. The experimental findings and statistical analysis show that when compared to the CV and GCV, our proposed quasi-oppositional learning pelican optimization algorithm provides greater performance in terms of computational time.

A Comparison between Some Penalized Methods for Estimating Parameters

Regression analysis is one of the most popular statistical methods in various biological and economic studies where, frequently, the number of explanatory variables becomes large. Penalized methods have been adapted and have gained popularity as a key for simultaneously performing variable selection and model estimation. This paper proposes contamination procedure from the viewpoint of different types of penalized regression, aiming at identifying any types of penalized methods that are best to deal with contamination data. This paper demonstrates that the Lasso regression is the best method for contamination data depending on the heavy tail distribution behavior of the response variables and using simulation for (15%) data with contamination. The comparison between types of penalized methods based on the statistical criterion (MAE and MSE) and results shows that the Lasso regression is better than another type of penalized method

Application of the generalized linear model to enable refractive index measurement with thermal sensitive interferometric sensors

In this work, the use of the generalized linear model (GLM) for refractive index detection, without dependence on temperature based on thermal sensitive interferometric arrangements is proposed. Here, the GLM was evaluated with four different link functions and the linear predictor was composed with a set of explanatory variables that represent different features of the interference pattern of the thermal sensitive interferometric arrangement. Besides, the refractive index sensor was modeled and experimentally implemented in order to support the technique. It is shown that the two models in which the sigmoidal and hyperbolic tangent link functions were used provided the best results. Moreover, it is shown that by using the stepwise regression technique, the number of explanatory variables can be reduced. Finally, for the experimental data, it is shown that by using the GLM with a sigmoidal function as a link function and a reduced set of 20 explanatory variables, the reached root mean square error and the determination coefficient were 0.0012 RIU and 0.9999, respectively for the refractive index range from 1.00 to 1.49 RIU.

FUNCTION GROUP SELECTION OF SEMBUNG LEAVES (BLUMEA BALSAMIFERA) SIGNIFICANT TO ANTIOXIDANTS USING OVERLAPPING GROUP LASSO

Functional groups of sembung leaf metabolites can be detected using FTIR spectrometry by looking at the spectrum's shape from specific peaks that indicate the type of functional group of a compound. There were 35 observations and 1866 explanatory variables (wavelength) in this study. The number of explanatory variables more than the number of observations is high-dimensional data. One method that can be used to analyze high-dimensional data is penalized regression. The overlapping group lasso method is a development of the group-based penalized regression method that can solve the problem of selecting variable groups and members of overlapping groups of variables. The results of selecting the variable groups using the overlapping group lasso method found that the functional groups that were significant for the antioxidants of sembung leaves were C=C Unstructured, CN amide, Polyphenol, Sio2.

Machine Learning Strategy for Improved Prediction of Micronutrient Concentrations in Soils of Taif Rose Farms Based on EDXRF Spectra

This study attempts to utilize newly developed machine learning techniques in order to develop a general prediction algorithm for agricultural soils in Saudi Arabia, specifically in the Taif region. Energy dispersive X-ray fluorescence (EDXRF) measurements were used to develop national predictive models that predict the concentrations of 14 micronutrients in soils of Taif rose farms, for providing high-quality data comparable to conventional methods. Machine learning algorithms used in this study included the simple linear model, the multivariate linear regression (MLR); and two nonlinear models, the random forest (RF) and multivariate adaptive regression splines (MARS). Our study proposes a machine learning (ML) strategy for predicting fertility parameters more accurately in agricultural soils using 10 farms of the Taif rose (Rosa damascena) in Taif, Saudi Arabia as a case study. Results demonstrated that MARS provides higher prediction performance when the number of explanatory variables is small, while RF is superior when the number of variables is large. On the other hand, the MLR is recommended as a moderate method for predicting multivariate variables. The study showed that multivariate models can be used to overwhelm the drawbacks of the EDXRF device, such as high detection limits and an element that cannot be directly measured.

Climate Extreme and Agriculture Development: Fresh Insight From Top Agri-Economics

A range of studies have been observed, covering the title of climate change and its linkage with the agriculture sector. This would justify the claim that changing environment has its several outcomes for which the agriculture sector cannot be ignored. The purpose of this study is to investigate the impact of various climate change dynamics and modelling on the four indicators of agriculture sector. Overall, five panel economies were selected having highest level of agriculture output in the world economy. The time duration of the study was during 1990–2018 with yearly data as collected from world development indicator or WDI. The study analysis was conducted while applying four panel regression models like ordinary least square, fixed effect estimator, least square dummy variable, and finally the random effect. For better understanding, study findings are empirically explained. The results confirm that both positive and negative impact of various proxies of climate change on agriculture dimension of selected economies. More specifically, it is observed that higher climate change in the form of carbon emission from different sources are causing a downturn effect on the agriculture export while at the same time, they are causing an up-ward shift in the agriculture import of selected economies. Besides, study has reasonably disrobed various policy implications both in theoretical and practical perspective. However, some limitations are also under observation. Firstly, this study considers the limited number of explanatory variables for reflecting the changing climate trends among top five agriculture economies of the world. However, there are still range of other factors which can be observed in the future studies to examine their influence on the selected indicators of agriculture industry. Secondly, this study has applied traditional panel models where no implication is observed for the dynamic panel methods like Generalized methods of Moments or GMM. Thirdly, this study has not provided any evidence for the cross-country analysis. Fourthly, this study has limited time span along with missing examination of both short run and long as well. Future studies may address these limitations for better implication in both theoretical and practical perspective.

Real-time detection of a change-point in a linear expectile model

In the present paper we address the real-time detection problem of a change-point in the coefficients of a linear model with the possibility that the model errors are asymmetrical and that the explanatory variables number is large. We build test statistics based on the cumulative sum (CUSUM) of the expectile function derivatives calculated on the residuals obtained by the expectile and adaptive LASSO expectile estimation methods. The asymptotic distribution of these statistics are obtained under the hypothesis that the model does not change. Moreover, we prove that they diverge when the model changes at an unknown observation. The asymptotic study of the test statistics under these two hypotheses allows us to find the asymptotic critical region and the stopping time, that is the observation where the model will change. The empirical performance is investigated by a comparative simulation study with other statistics of CUSUM type. Two examples on real data are also presented to demonstrate its interest in practice.

Performance attribution for multifactorial equity portfolios

This paper revisits the cross-sectional approach to the performance analysis of multifactor investment strategies. Its main contributions are threefold: first, the use of a cross-sectional projection of asset returns onto the factor portfolio weights to form approximate portfolio returns; second, the introduction of nonlinear interaction terms between factors that faithfully reproduce the investment portfolio construction; and third, a natural and intuitive decomposition of the portfolio performance as the sum of factor contributions. The method we propose has several advantages over other time-series-based or general cross-sectional regression models: it faithfully reflects the current state of the investment portfolio, it is parsimonious in the number of explanatory variables, it leads to an approximation of the portfolio returns that has a small residual error and it provides a straightforward interpretation of the portfolio performances in terms of the factors it is designed from. The method we advocate is first presented and explained in detail, and then concrete applications to multifactor equity strategies are presented.

Modeling of discrete questionnaire data with dimension reduction

The paper deals with the task of modeling discrete questionnaire data with a reduced dimension of the model. The discrete model dimension is reduced using the construction of local models based on independent binomial mixtures estimated with the help of recursive Bayesian algorithms in the combination with the naive Bayes technique. The main contribution of the paper is a three-phase algorithm of the discrete model dimension reduction, which allows to model high-dimensional questionnaire data with high number of explanatory variables and their possible realizations. The proposed general solution is applied to the traffic accident questionnaire analysis, where it takes the form of the classification of the accident circumstances and prediction of the traffic accident severity using the currently measured discrete data. Results of testing the obtained model on real data and comparison with theoretical counterparts are demonstrated.

Objective Bayesian variable selection in linear regression model

Variable selection in a regression model with k potential explanatory variables requires the choosing of a model among the possible submodels, which is a difficult task when the number of explanatory variables is moderately large. In this study, we propose the objective Bayesian variable selection procedures where the encompassing of the underlying nonnested linear models is crucial. Based on the encompassed models, objective priors for the multiple testing problem involved in the variable selection problem can be defined. The proposed approach provides a considerable reduction in the size of the compared models by restricting the posterior search for the right models, from to only k + 1, given k explanatory variables. Furthermore, the consistency of the proposed variable selection procedures was checked and their performance was examined using real examples and simulation analyzes by comparing the classical and Bayesian procedures of search in all possible submodels.

The impact of latent topic valence of online reviews on purchase intention for the accommodation industry

This study utilizes machine learning (ML) natural language processing (NLP) algorithms and statistical methods in order to measure the impact that qualitative textual reviews have on booking intentions for accommodations. Using over 400,000 online reviews from 1256 accommodations in South Korea, latent Dirichlet allocation (LDA) is used to determine the topic of the review content, convolutional neural networks (CNNs) are used to identify the valence of the reviews, and spatial probit models are used to determine the impact of the review content and valence on booking intention, while controlling for several other variables. It is found that positive reviews about an accommodation’s ambiance, value, service, front office, accessibility, surrounding neighborhood and room capacity result in significantly higher booking intentions, while negative reviews in the service, front office and surrounding neighborhood result in lower probability of booking. A number of explanatory variables also have varying effects on booking intentions.

Explaining changes in perceived wildfire risk related to the mountain pine beetle outbreak in north central Colorado

Longitudinal studies of risk perception, while growing, remain an understudied area of risk analysis research. Natural resource-based communities provide a key backdrop for analyzing dynamic risk perceptions and related social-ecological processes. Since the mid-1990s, a mountain pine beetle (MPB) outbreak has affected roughly 3.4 million acres of north central Colorado forests. Slow-moving landscape changes have taken place as a result of the outbreak, making it a relevant case study in which to examine wildfire and forest risk perceptions over time. In 2018, we replicated and expanded a 2007 mail survey of residents about local response to the MPB outbreak in nine north central Colorado communities. Using the initial and follow-up survey data from these study communities, we examined the changing wildfire risk perception and its relationships with various aspects of local human–nature interactions. The longitudinal surveys showed that perceptions of all forest risks except for forest fire and falling trees declined across the two study phases. Respondents in the 2018 survey also indicated increases in perceived likelihood of a wildfire, perceived severity of possible wildfire damage, and general concern about wildfire hazard. The analysis suggested that evolving wildfire risk perception was strongly associated with a number of explanatory variables and personal characteristics. Higher levels of perceived tree mortality, community interaction and communication, and individual actions in response to forest risks were consistently related to greater odds of indicating increased wildfire risk perception. In general, community social vulnerability, perceived tree re-growth, satisfaction with local management entities, and age were negatively related to the likelihood of having elevated perceptions of fire risk, whereas males and forest-related occupations were associated with relatively lower odds of reporting increases in perceived wildfire risk. Significant influencing factors for change also varied across the cognitive and affective dimensions of wildfire risk perception. Findings of this research have important implications for conceptual approaches, management and policy, as well as social-ecological indication and analysis in forest risk related areas.

An Modified PLSR Method in Prediction

Among many statistical methods for linear models with the multicollinearity problem, partial least squares regression (PLSR) has become, in recent years, increasingly popular and, very often, the best choice. However, while dealing with the predicting problem from automobile market, we noticed that the results from PLSR appear unstable though it is still the best among some standard statistical methods. This unstable feature is likely due to the impact of the information contained in explanatory variables that is irrelevant to the response variable. Based on the algorithm of PLSR, this paper introduces a new method, modified partial least squares regression (MPLSR), to emphasize the impact of the relevant information of explanatory variables on the response variable. With the MPLSR method, satisfactory predicting results are obtained in the above practical problem. The performance of MPLSR, PLSR and some standard statistical methods are compared by a set of Monte Carlo experiments. This paper shows that the MPLSR is the most stable and accurate method, especially when the ratio of the number of observation and the number of explanatory variables is low.

Discovery of Depression-Associated Factors From a Nationwide Population-Based Survey: Epidemiological Study Using Machine Learning and Network Analysis

BackgroundIn epidemiological studies, finding the best subset of factors is challenging when the number of explanatory variables is large.ObjectiveOur study had two aims. First, we aimed to identify essential depression-associated factors using the extreme gradient boosting (XGBoost) machine learning algorithm from big survey data (the Korea National Health and Nutrition Examination Survey, 2012-2016). Second, we aimed to achieve a comprehensive understanding of multifactorial features in depression using network analysis.MethodsAn XGBoost model was trained and tested to classify “current depression” and “no lifetime depression” for a data set of 120 variables for 12,596 cases. The optimal XGBoost hyperparameters were set by an automated machine learning tool (TPOT), and a high-performance sparse model was obtained by feature selection using the feature importance value of XGBoost. We performed statistical tests on the model and nonmodel factors using survey-weighted multiple logistic regression and drew a correlation network among factors. We also adopted statistical tests for the confounder or interaction effect of selected risk factors when it was suspected on the network.ResultsThe XGBoost-derived depression model consisted of 18 factors with an area under the weighted receiver operating characteristic curve of 0.86. Two nonmodel factors could be found using the model factors, and the factors were classified into direct (P<.05) and indirect (P≥.05), according to the statistical significance of the association with depression. Perceived stress and asthma were the most remarkable risk factors, and urine specific gravity was a novel protective factor. The depression-factor network showed clusters of socioeconomic status and quality of life factors and suggested that educational level and sex might be predisposing factors. Indirect factors (eg, diabetes, hypercholesterolemia, and smoking) were involved in confounding or interaction effects of direct factors. Triglyceride level was a confounder of hypercholesterolemia and diabetes, smoking had a significant risk in females, and weight gain was associated with depression involving diabetes.ConclusionsXGBoost and network analysis were useful to discover depression-related factors and their relationships and can be applied to epidemiological studies using big survey data.

Multiple model triangulation to identify factors associated with lameness in British sheep flocks

Identification of factors associated with an outcome can be challenging when the number of explanatory variables is large in relation to the number of observations. Multiple model triangulation, where results from several model types are combined, improves the likelihood of identifying true predictor variables. The aim of this study was to use triangulation to identify covariates likely to be truly associated with the prevalence of lameness in sheep flocks in Great Britain.Data were collected using a questionnaire sent to 3200 sheep farmers in Great Britain in 2018. The useable response rate was 14.1 %. The geometric mean prevalence of lameness was 1.4 % (95 % CI 1.2−1.7) for ewes, and 0.6 % (95 % CI 0.5−0.9) for lambs, however, approximately 60 % flocks had >2% prevalence of lameness in ewes.Four model types were investigated, two generalised linear models (negative binomial and quasi-Poisson) built using stepwise selection, and two elastic net models (Poisson and Gaussian distributions) refined with selection stability estimation.Triangulated covariates were those selected in three or all four models – 10 for ewes and 12 for lambs. Higher prevalence of lameness in ewes was associated with 5−100% feet bleeding during routine foot trimming compared with not foot trimming, footbathing the flock to treat severe footrot (SFR) and always using formalin in footbaths, both compared with not footbathing, using FootVax™ for <1 year compared with not using FootVax™, and never quarantining new or returning sheep to the farm for >3 weeks compared with always. Lower prevalence of lameness in ewes was associated with vaccinating with FootVax™ for >5 years compared with not vaccinating, peat soil compared with no peat soil, and having no lame ewes to treat.Higher prevalence of lameness in lambs was associated with 5−100% feet bleeding during routine foot trimming, always foot trimming ewes with SFR, not knowingly selecting replacement ewes from ewes that were never lame compared with always, replacement sheep purchased and homebred compared with only homebred, treating lambs >3 days after recognition of lameness compared with 0-3 days and footbathing the flock to treat interdigital dermatitis compared with not footbathing at all. Lower prevalence of lameness in lambs was associated with peat soil, flocks in Scotland versus England, an altitude of >230−500 m compared with ≤230 m, never using antibiotic injection to treat lambs with SFR compared with always, and having no lame lambs to treat.We conclude triangulation identified reliable management practices for farmers to implement to minimise lameness in sheep.

Progressive familial socialization and white partisans’ racial attitudes

ABSTRACT Scholars have correlated the racial attitudes of White partisans with a number of explanatory variables, including ingroup favoritism and outgroup prejudice. Notwithstanding the importance of these variables, scholars have neglected other constructs that may also be relevant to understanding White racial attitudes. This paper addresses this important lacuna by analyzing how the racial attitudes of White partisans are shaped by progressive familial socialization (PFS) – defined as the frequency that a White individual was socialized by their families to be aware of the structural advantages of Whiteness. Drawing on data from Wave 43 of the American Trends Panel, the findings demonstrate that PFS is associated with increased awareness of the structural disadvantages faced by Blacks, and that these effects are more salient for White Democrats than White Republicans. I also find that PFS is politically consequential, with Whites socialized progressively on race being more likely to think that Trump has made race relations worse since taking office. More importantly, these effects may cut across partisan evaluations of Trump’s handling of race relations. I conclude that PFS is a construct that has significant implications for understanding what familial socialization practices on race may do for the future of US race relations.

Generalized two-parameter estimators in the multinomial logit regression model: methods, simulation and application

In this article, we propose generalized two-parameter (GTP) estimators and an algorithm for the estimation of shrinkage parameters to combat multicollinearity in the multinomial logit regression model. In addition, the mean squared error properties of the estimators are derived. A simulation study is conducted to investigate the performance of proposed estimators for different sample sizes, degrees of multicollinearity, and the number of explanatory variables. Swedish football league dataset is analyzed to show the benefits of the GTP estimators over the traditional maximum likelihood estimator (MLE). The empirical results of this article revealed that GTP estimators have a smaller mean squared error than the MLE and can be recommended for practitioners.