Neural network can beat conventional forecasts – Kansas City Fed research


Machine learning techniques can overcome many of the shortcomings in conventional forecasting, researchers at the Federal Reserve Bank of Kansas City say in a new working paper.

Thomas Cook and Aaron Hall use a neural network – a form of “deep learning” – to generate a forecast model while remaining “agnostic” to functional form. They test four different architectures and find the encoder-decoder form, originally designed for language modelling, works best.

Each of the four methods produce

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact [email protected] or view our subscription options here:

You are currently unable to copy this content. Please contact [email protected] to find out more.

To continue reading...

You need to sign in to use this feature. If you don’t have a Central Banking account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: