In this paper, the authors discuss the method of Bayesian regression and its efficacy for predicting price variation of Bitcoin. This regression refers to utilising empirical data as proxy to
perform Bayesian inference. They utilise Bayesian regression for the so-called “latent source model”.

The Bayesian regression for “latent source model” was introduced and discussed by Chen, Nikolov and Shah and Bresler, Chen and Shah for the purpose of binary classification. They established theoretical as well as empirical efficacy of the method for the setting of binary classification. In this paper, it is instead utilised for predicting real-valued quantity, the price of Bitcoin. Based on this price prediction method, the authors devise a simple strategy for trading Bitcoin. The strategy is able to nearly double the investment in less than 60 day period when run against real data trace.

Download this paper to access a variety of different problems, the classical approach and an alternative approach offered by the authors.