A1004
Title: Adaptive forecasting: Implementation (R package forecastADAPT)
Authors: Liudas Giraitis - Queen Mary University of London (United Kingdom) [presenting]
Violetta Dalla - National and Kapodistrian University of Athens (Greece)
George Kapetanios - Kings College London (United Kingdom)
Abstract: Forecasting strategies that are robust to structural breaks and structural changes have earned renewed attention in the literature. These strategies are particularly attractive to applied econometricians because they are built on weighted averages of down-weighted past information. They are easy to implement, and in general, they are robust to different types of structural change: they are capable of adapting to it in real time. The aim is to introduce the R package forecastADAPT that implements the robust forecasting strategies via a set of examples and applications that might be of interest to applied econometricians. The theory behind this R package, how the forecast is constructed, the forecast error estimated, and the data-based tuning parameter selected, is based on a prior study.
