TY - JOUR
T1 - Log-symmetric quantile regression models
AU - Saulo, Helton
AU - Dasilva, Alan
AU - LEIVA SANCHEZ, VICTOR ELISEO
AU - Sánchez, Luis
AU - DE LA FUENTE MELLA, HANNS ANIBAL
N1 - Funding Information:
The authors thank the Editors and Reviewers for their constructive comments on an earlier version of this manuscript. The research of H. Saulo and A. Dasilva was partially supported by the National Council for Scientific and Technological Development (CNPq) and the Coordination for the Improvement of Higher Education Personnel (CAPES) from the Brazilian federal government under the Ministry of Science and Technology; the research of V. Leiva and L. Sánchez was partially supported by grant FONDECYT 1200525 from the National Agency for Research and Development (ANID) of the Chilean government under the Ministry of Science, Technology, Knowledge and Innovation; and the research of H. de la Fuente‐Mella was partially supported by grant Núcleo de Investigación en Data Analytics/VRIEA/PUCV/039.432/2020 from the Vice‐Rectory for Research and Advanced Studies of the Pontifical Catholic University of Valparaíso, Chile.
Publisher Copyright:
© 2021 Netherlands Society for Statistics and Operations Research
PY - 2021
Y1 - 2021
N2 - Regression models based on the log-symmetric family of distributions are particularly useful when the response variable is continuous, positive, and asymmetrically distributed. In this article, we propose and derive a class of models based on a new approach to quantile regression using log-symmetric distributions parameterized by means of their quantiles. Two Monte Carlo simulation studies are conducted utilizing the R software. The first one analyzes the performance of the maximum likelihood estimators, the Akaike, Bayesian, and corrected Akaike information criteria, and the generalized Cox–Snell and random quantile residuals. The second one evaluates the size and power of the Wald, likelihood ratio, score, and gradient tests. A web-scraped box-office data set of the movie industry is analyzed to illustrate the proposed approach. Within the main results of the simulation carried out, the good performance of the maximum likelihood estimators is reported.
AB - Regression models based on the log-symmetric family of distributions are particularly useful when the response variable is continuous, positive, and asymmetrically distributed. In this article, we propose and derive a class of models based on a new approach to quantile regression using log-symmetric distributions parameterized by means of their quantiles. Two Monte Carlo simulation studies are conducted utilizing the R software. The first one analyzes the performance of the maximum likelihood estimators, the Akaike, Bayesian, and corrected Akaike information criteria, and the generalized Cox–Snell and random quantile residuals. The second one evaluates the size and power of the Wald, likelihood ratio, score, and gradient tests. A web-scraped box-office data set of the movie industry is analyzed to illustrate the proposed approach. Within the main results of the simulation carried out, the good performance of the maximum likelihood estimators is reported.
KW - econometric models
KW - hypothesis testing
KW - log-symmetric distributions
KW - R software
KW - web scraping
UR - http://www.scopus.com/inward/record.url?scp=85104950173&partnerID=8YFLogxK
U2 - 10.1111/stan.12243
DO - 10.1111/stan.12243
M3 - Article
AN - SCOPUS:85104950173
JO - Statistica Neerlandica
JF - Statistica Neerlandica
SN - 0039-0402
ER -