Penalised regressions vs. autoregressive moving average models for forecasting inflation

This article relates the Seasonal Autoregressive Moving Average Models (SARMA) to linear regression. Based on this relationship, the paper shows that penalized linear models can outperform the out-of-sample forecast accuracy of the best SARMA models in forecasting inflation as a function of past val...

Full description

Autores:: Ospina-Holguín, Javier Humberto
Padilla Ospina, Ana Milena

Tipo de recurso:: Article of journal

Fecha de publicación:: 2019

Institución:: Corporación Universidad de la Costa

Repositorio:: REDICUC - Repositorio CUC

Idioma:: eng

Description
Summary:	This article relates the Seasonal Autoregressive Moving Average Models (SARMA) to linear regression. Based on this relationship, the paper shows that penalized linear models can outperform the out-of-sample forecast accuracy of the best SARMA models in forecasting inflation as a function of past values, due to penalization and cross-validation. The paper constructs a minimal functional example using edge regression to compare both competing approaches to forecasting monthly inflation in 35 selected countries of the Organization for Economic Cooperation and Development and in three groups of coun-tries. The results empirically test the hypothesis that penalized linear regression, and edge regression in particular, can outperform the best standard SARMA models calculated through a grid search when fore-casting inflation. Thus, a new and effective technique for forecasting inflation based on past values is provided for use by financial analysts and investors. The results indicate that more attention should be paid to automatic learning techniques for forecasting inflation time series, even as basic as penalized linear regressions, because of their superior empirical performance.

Penalised regressions vs. autoregressive moving average models for forecasting inflation

Publicaciones similares