Mixtures of quantile regressions

Wu, Qiang; Yao, Weixin

doi:10.1016/j.csda.2014.04.014

Cited by 19 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, Data Science has several times been the focus of different debates with the purpose of defining its distinction (or its standardization) with respect to Statistics. 4 Just to mention one instance, C. F. Jeff Wu, during his inaugural lecture for the H. C. Carver Professorship in Statistics at the University of Michigan in 1997, 5 claimed that Statistics should be renamed Data Science and Statisticians Data Scientists. The new modern methodologies, however, are pooling the two disciplines of statistics and computer science as in the interaction of computational algorithms with cognitive science in artificial intelligence and the viewpoint of machine learning as a marriage of statistics and knowledge representation.…”

Section: Data Science: Origin and Developmentmentioning

confidence: 99%

A review of data science in business and industry and a future view

Vicario

Coleman

2019

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

The aim of this paper is to frame Data Science, a fashion and emerging topic nowadays in the context of business and industry. We open with a discussion about the origin of Data Science and its requirement for a challenging mix of capability in data analytics, information technology, and business know‐how. The mission of Data Science is to provide new or revised computational theory able to extract useful information from the massive volumes of data collected at an accelerating pace. In fact, besides the traditional measurements, digital data obtained from images, text, audio, sensors, etc complement the survey. Then, we review the different and most popular methodologies among the practitioners of Data Science research and applications. In addition, because the emerging field requires personnel with new competences, we attempt to describe the Data Scientist profile, one of the sexiest jobs of the 21st Century according to Davenport and Patil. Most people are aware of the need to embrace Data Science, but they feel intimidated that they do not understand it and they worry that their jobs will disappear. We want to encourage them: Data Science is more likely to add value to jobs and enrich the lives of working people by helping them make better, more informed business decisions. We conclude this paper by presenting examples of Data Science in action in business and industry, to demonstrate the collection of specialist skills that must come together for this new science to be effective.

show abstract

Section: Data Science: Origin and Developmentmentioning

confidence: 99%

A review of data science in business and industry and a future view

Vicario

Coleman

2019

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

show abstract

“…The identifiability of the proposed model has been established by Wang et al (2012), Balabdaoui & Doss (2014), and Wu & Yao (2016). We propose two EM-type algorithms, which aim at adjusting the model misspecification.…”

Section: Introductionmentioning

confidence: 99%

“…Possible solutions include traditional nonparametric methods, e.g. Hunter & Young (2012) and Wu & Yao (2016), to adjust the parametric model mis-specification. These traditional nonparametric methods, e.g.…”

Section: Introductionmentioning

confidence: 99%

The robust EM-type algorithms for log-concave mixtures of regression models

Yao

2017

Computational Statistics & Data Analysis

Self Cite

View full text Add to dashboard Cite

Finite mixture of regression (FMR) models can be reformulated as incomplete data problems and they can be estimated via the expectation-maximization (EM) algorithm. The main drawback is the strong parametric assumption such as FMR models with normal distributed residuals. The estimation might be biased if the model is misspecified. To relax the parametric assumption about the component error densities, a new method is proposed to estimate the mixture regression parameters by only assuming that the components have log-concave error densities but the specific parametric family is unknown. Two EM-type algorithms for the mixtures of regression models with log-concave error densities are proposed. Numerical studies are made to compare the performance of our algorithms with the normal mixture EM algorithms. When the component error densities are not normal, the new methods have much smaller MSEs when compared with the standard normal mixture EM algorithms. When the underlying component error densities are normal, the new methods have comparable performance to the normal EM algorithm.

show abstract

“…On the one hand, statisticians paid attention to improving the robustness of mixture regression model, [6] used the t-distribution for overcoming the influence of outliers and [7] introduced a robust mixture regression model by assuming the error terms follow a Laplace distribution. Further, Wu et al [8] dropped any parametric assumption about the error densities and proposed the mixture of quantile regressions model. On the other hand, variable selection became a research hotspot in mixture regression modeling.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Sparse Group Variable Selection for a Robust Mixture Regression Model Based on Laplace Distribution

Wang¹,

Ye²

2020

APM

View full text Add to dashboard Cite

The traditional estimation of Gaussian mixture model is sensitive to heavytailed errors; thus we propose a robust mixture regression model by assuming that the error terms follow a Laplace distribution in this article. And for the variable selection problem in our new robust mixture regression model, we introduce the adaptive sparse group Lasso penalty to achieve sparsity at both the group-level and within-group-level. As numerical experiments show, compared with other alternative methods, our method has better performances in variable selection and parameter estimation. Finally, we apply our proposed method to analyze NBA salary data during the period from 2018 to 2019.

show abstract

Mixtures of quantile regressions

Cited by 19 publications

References 28 publications

A review of data science in business and industry and a future view

A review of data science in business and industry and a future view

The robust EM-type algorithms for log-concave mixtures of regression models

Adaptive Sparse Group Variable Selection for a Robust Mixture Regression Model Based on Laplace Distribution

Contact Info

Product

Resources

About