A working paper published by the Bank of Spain presents a number of methods for averaging the predictive results of non-linear models, and applies them to data from the US.
In Model averaging in Markov-switching models: predicting national recessions with regional data, Pierre Guérin and Danilo Leiva-Leon introduce what they call “new weighting schemes to combine discrete forecasts from competing Markov-switching models”.
A large literature combines results from linear forecasting models, but
- Latvian police detain central bank governor on corruption charges
- RBI under pressure from alleged banking fraud scandal
- Ireland withdraws Lane’s nomination for ECB vice president
- Saudi Arabia’s Alkholifey on economic restructuring, reserves and cyber security
- Central banks mull securities lending as response to low yields