Generalised density forecast combinations

Working papers set out research in progress by our staff, with the aim of encouraging comments and debate.
Published on 28 March 2014

Working Paper No. 492
By Nicholas Fawcett, George Kapetanios, James Mitchell and Simon Price

Density forecast combinations are becoming increasingly popular as a means of improving forecast ‘accuracy’, as measured by a scoring rule. In this paper we generalise this literature by letting the combination weights follow more general schemes. Sieve estimation is used to optimise the score of the generalised density combination where the combination weights depend on the variable one is trying to forecast. Specific attention is paid to the use of piecewise linear weight functions that let the weights vary by region of the density. We analyse these schemes theoretically, in Monte Carlo experiments and in an empirical study. Our results show that the generalised combinations outperform their linear counterparts. 

PDF Generalised density forecast combinations 

Was this page useful?
Add your details...