Forecasting Spanish unemployment with Google Trends and dimension reduction techniques

This paper presents a method to improve the one-step-ahead forecasts of the Spanish unemployment monthly series. To do so, we use a large number of potential explanatory variables extracted from searches in Google (Google Trends tool). Two different dimension reduction techniques are implemented to...

ver descrição completa

Detalhes bibliográficos
Autores: Mulero, Rodrigo, García Hiernaux, Alfredo Alejandro
Formato: artículo
Fecha de publicación:2020
País:España
Recursos:Universidad Complutense de Madrid (UCM)
Repositorio:Docta Complutense
Idioma:inglés
OAI Identifier:oai:docta.ucm.es:20.500.14352/6356
Acesso em linha:https://hdl.handle.net/20.500.14352/6356
Access Level:acceso abierto
Palavra-chave:C32
C52
C53
Unemployment
Forecasting
Google Trends
Dimensionality reduction
RMSE
Economía
Econometría (Economía)
Indicadores económicos
53 Ciencias Económicas
5302 Econometría
5302.01 Indicadores Económicos
Descrição
Resumo:This paper presents a method to improve the one-step-ahead forecasts of the Spanish unemployment monthly series. To do so, we use a large number of potential explanatory variables extracted from searches in Google (Google Trends tool). Two different dimension reduction techniques are implemented to decide how to combine the explanatory variables or which ones to use. The results reveal an increase in predictive accuracy of 10-25%, depending on the dimension reduction method employed. A deep robustness analysis confirms this findings, as well as the relevance of using a large amount of Google queries together with a dimension reduction technique, when no prior information on which are the most informative queries is available.