Publication

Abstract : Bitcoin has attracted considerable attention in today’s world because of the combination of encryption technology along with the monetary units. For traders, Bitcoin leads to a promising investment because of its highly fluctuating price. Block chain technology assists in the transactions of documentation. The characteristics of the bitcoin which is derived from the blockchain technology has led to diverse interests in the field of economics. The bitcoin data is selected from 2013 to 2018, over a period of 5 years for this analysis. Here a new roll over technology is applied where new data is obtained over time which will close out the old information during machine training. This mechanism will help in incorporating new information in the short-term learning. The results show that the rollover mechanism improves the time series prediction accuracy.