GlobalSearchRegression.jl: Building bridges between Machine Learning and Econometrics in Fat-Data scenarios

The aim of this paper is twofold. The first one is to describe a novel research-project designed for building bridges between machine learning and econometric worlds (ModelSelection.jl).The second one is to introduce the main characteristics and comparative performance of the first Julia-native all-subset regression algorithm included in GlobalSearchRegression.jl (v1.0.5). As other available alternatives, this algorithm allows researchers to obtain the best model specification among all possible covariate combinations - in terms of user defined information criteria-, but up to 3165 and 197 times faster than STATA and R alternatives, respectively.

Palabras clave: PARALLEL COMPUTING , ECONOMETRICS , MACHINE LEARNING , FAT-DATA

Ver el registro completo

Archivos asociados

Tamaño: 276.0Kb

Formato: PDF

Descargar

Licencia

Excepto donde se diga explícitamente, este item se publica bajo la siguiente descripción: Creative Commons Attribution 2.5 Unported (CC BY 2.5)

Identificadores

URI: http://hdl.handle.net/11336/211320

DOI: http://dx.doi.org/10.21105/jcon.00053

URL: https://proceedings.juliacon.org/papers/10.21105/jcon.00053

Colecciones

Articulos(CCT - LA PLATA)
Articulos de CTRO.CIENTIFICO TECNOL.CONICET - LA PLATA

Citación

Panigo, Demian Tupac; Gluzmann, Pablo Alfredo; Mocskos, Esteban Eduardo; Mauri Ungaro, Adán; Mari, Valentin; et al.; GlobalSearchRegression.jl: Building bridges between Machine Learning and Econometrics in Fat-Data scenarios; Juliacon; Proceedings of the JuliaCon Conferences; 2; 53; 6-2020; 1-6

Altmétricas