eduzhai > Applied Sciences > Engineering >

Mean Reversion: A New Approach

  • Save

... pages left unread,continue reading

Document pages: 13 pages

Abstract: In this paper, we review briefly existing approaches to statistical arbitrage and mean reversion and then proceed to present a new approach that combines model-independence, through the use of neural networks, with information theory. The neural networks parameters were calibrated in a way that guarantees mean reversion, using a loss function that aims to maximise the entropy of the power spectrum of the error instead of the traditionally minimised mean squared error. The returns of a signal-based strategy derived from this new approach were empirically illustrated using historical stocks data from several US firms. The results showed significant gains can be realised by using this new approach.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×