BITCOIN ANALYSIS AND PREDICTION

Following the recent boom and bust of cryptocurrency prices, Bitcoin is increasingly regarded as an investment asset. Because of its extreme volatility, there is a need for accurate forecasts on which to make investment decisions. Although previous research has used machine learning to predict Bitcoin prices more accurately, few studies have focused on the feasibility of applying different modelling techniques to samples with different data structures and dimensional features. To anticipate Bitcoin price at various frequencies using machine learning techniques, we first divide Bitcoin price into two categories: daily price and high-frequency pricing. For Bitcoin daily price prediction, a collection of high-dimension features such as property and network, trade and market, attention, and gold spot price are employed, whereas basic trading features obtained from a cryptocurrency exchange are used for 5-minute interval price prediction. Deep learning models lstm and bi-directional lstm are used. Because bitcoin has grown dramatically in recent years, it is critical to forecast its price more accurately because it will be a future investment trend. People will be able to invest more effectively if prices are forecasted more accurately.