Predictivistic Bayesian Forecasting System

National Bank of Poland Working Paper No. 87

27 Pages Posted: 8 Jul 2011

See all articles by Andrzej Kociecki

Andrzej Kociecki

National Bank of Poland

Marcin Kolasa

National Bank of Poland; Warsaw School of Economics (SGH)

Michał Rubaszek

National Bank of Poland; Warsaw School of Economics (SGH)

Date Written: 2011

Abstract

This paper introduces a formal method of combining expert and model density forecasts when the sample of past forecasts is unavailable. It works directly with the expert forecast density and endogenously delivers weights for forecast combination, relying on probability rules only. In the empirical part, we illustrate how our framework can be applied in forecasting US inflation by mixing density forecasts from an autoregressive model and the Survey of Professional Forecasters.

Keywords: Combining density forecasts, Forecast evaluation, Bayesian inference, Predictivism

JEL Classification: C11, C53

Suggested Citation

Kociecki, Andrzej and Kolasa, Marcin and Rubaszek, Michal, Predictivistic Bayesian Forecasting System (2011). National Bank of Poland Working Paper No. 87. Available at SSRN: https://ssrn.com/abstract=1881644 or http://dx.doi.org/10.2139/ssrn.1881644

Andrzej Kociecki (Contact Author)

National Bank of Poland ( email )

00-919 Warsaw
Poland

Marcin Kolasa

National Bank of Poland ( email )

00-919 Warsaw
Poland

Warsaw School of Economics (SGH) ( email )

aleja Niepodleglosci 162
PL-Warsaw, 02-554
Poland

Michal Rubaszek

National Bank of Poland ( email )

00-919 Warsaw
Poland

Warsaw School of Economics (SGH) ( email )

aleja Niepodleglosci 162
PL-Warsaw, 02-554
Poland

Register to save articles to
your library

Register

Paper statistics

Downloads
32
Abstract Views
446
PlumX Metrics