ARCHIVES

Cryptocurrencies as distributed digital assets (commonly in Bitcoin, BTC, Ethereum, ETH, Ripple, XRP and Litecoin, LTC) have infiltrated the financial markets, and decentralized transactions take place on blockchain block structures. They are highly volatile in that they can rise and fall at any given time and this is both beneficial and demerits as far as the gains and losses are concerned. Our project is based on the concepts of Long Short Term Memory (LSTM) networks, in which we create a deep learning predictive model, which can predict the future prices of these cryptocurrencies and that can be used by the market participants to take well-informed decisions. Market Momentum Trading Strategies: These models are capable of establishing sequential information tendencies of a price over the years and can very successfully project future results according to the manner in which a price has been acting in the past years. In this project, the price history of Bitcoin, Ethereum, Ripple and Litecoin are obtained in the Binance API and pre-processed. The LSTM model is then trained to memorise the trends of the sequential data of the Close price and future forecast of the prices is then made based on the past trends. Normalization, train test split and reorganization of the data based on LSTM architecture The performance will be measured in meansquared error(MSE limit) and loss. The model can be implemented to each of the four cryptocurrencies and offers the flexibility of predictions, depending on the choice of the user. The comparison between the predictive prices and the actual prices is carried out and the results are plotted in an attempt to justify the accuracy of the model. The system gives a more sophisticated way of learning the volatility of the cryptocurrency markets using the techniques of ML/DL, which will also be beneficial in providing the investors and the trader at large with the insights they will need in the future in terms of predicting the future price trends and in terms of risk management.