Compositional combination and selection of forecasters

The Split-Then-Combine approach has previously been used to generate the weights of forecasts in a combination in the Euclidean space. This paper extends this approach to combine forecasts inside the simplex space, the sample space of positive weights adding up to one. As it turns out, the simplicia...

ver descrição completa

Detalhes bibliográficos
Autores: Martín Arroyo, Antonio S., Juan Fernández, Aranzazu de
Formato: artículo
Fecha de publicación:2022
País:España
Recursos:Universidad Autónoma de Madrid
Repositorio:Biblos-e Archivo. Repositorio Institucional de la UAM
Idioma:inglés
OAI Identifier:oai:repositorio.uam.es:10486/712483
Acesso em linha:http://hdl.handle.net/10486/712483
https://dx.doi.org/10.2436/20.8080.02.123
Access Level:acceso abierto
Palavra-chave:Aitchison geometry
Combination-After-Selection
Dimensionality problem
Simplex
Split-Then-Combine
Economía
Descrição
Resumo:The Split-Then-Combine approach has previously been used to generate the weights of forecasts in a combination in the Euclidean space. This paper extends this approach to combine forecasts inside the simplex space, the sample space of positive weights adding up to one. As it turns out, the simplicial statistic given by the sample centre compares favourably against the fxed-weight, average forecast. Besides, we also develop a Combination-After-Selection method to get rid of redundant forecasters. We apply these approaches to make out-of-sample one-step ahead combinations and subcombinations of forecasts for several economic variables. This methodology is particularly useful when the sample size is smaller than the number of forecasts, a case where other methods (e.g., ordinary least squares or principal component analysis) are not applicable