Forecasting time series data is an important subject in economics, business, and finance. The paper Forecasting Economic and Financial Time Series: ARIMA vs. LSTM advocates the benefits of applying deep learning-based algorithms to the economics and financial data. Our work in this report is evaluating the merits of the paper’s comparative approach to predicting financial time series using LSTM vs. ARIMA models by applying both models on common financial time series. We will also evaluate alternative deep-learning models on more fine-grained Bitcoin data. Experimental results show that following the procedure of the paper does not generate the result that supports the superiority of the LSTM model over ARIMA model; and our trading strategy on Bitcoin using LSTM model generates slight better return than Bitcoin return over the same period, but the trading strategy is highly dependent on the chosen level of trading aggressiveness and on our chosen hyperparameters. Keywords: Bitcoin; trading strategy; crypotocurrency; neural networks; Forecasting; Economic and Financial Time Series; Dee

⚠ No verified equity curve — no track-record source connected.

No score history is stored yet — only the current score is shown.

⚠ No live verification account connected — ask for proof before buying.

Alerts on changes: coming soon

Prop-firm compatibility not provided.

Data-completeness & trust index (not a profitability rating)