Mehrad Mashoof - Department of Industrial Engineering, Science and Research branch, Islamic Azad University, Tehran, Iran
Abbas Saghaei - Department of Industrial Engineering, Science and Research branch, Islamic Azad University, Tehran, Iran
Amir Azizi - assistant professor of industrial engineering department, engineering faculty, science and research branch, Islamic Azad University, Tehran

It might be challenging to manage a portfolio in the cryptocurrency market. It can be difficult to choose from among thousands of assets, forecast costs, gauge returns, and assess risks. In this work, we use an LSTM deep learning model to forecast the return for each chosen cryptocurrency. Furthermore, we integrated the predicted return with three conventional portfolio optimization methods, namely MV, SHARPE, and NAVE, to demonstrate the superiority of our methodology in terms of portfolio return factors. The assessment is based on historical data for the ۱۲ months from January ۱, ۲۰۲۱, to December ۳۱, ۲۰۲۱. The results of the experiments demonstrate that our robust portfolio optimization outperforms conventional methods in terms of the portfolio return criteria.

Portfolio Optimization, Cryptocurrency Trading, Return Forecasting, Deep Learning, Robust Optimization