This repository demonstrates a practice project in the price forecasting field using data from a Kaggle dataset - Cryptocurrency Historical Prices.
dash folder: this folder contains the python scripts and assets for the dash app.
notebooks folder: this folder contains two notebooks, EDA.ipynb and Price_Forecast.ipynb.
The EDA.ipynb contains the explotary data analysis and data visualization.
The Price_Forecast.ipynb contains the price forecast analysis for Bitcoin, in which I demonstrated three different models / approaches.
- the time series model (ARIMA),
- the machine learning models (random forest, gradient boosting, xgboost),
- the deep learning model.
Screenshot 1: Daily close price series of top 10 cryptocurrencies by market value
Screenshot 2: Moving average and bollinger band plot of selected cryptocurrecy (by clicking the corresponding pie slide)
Screenshot 3: Moving average and bollinger band plot of selected cryptocurrecy (over selected time period)
Screenshot 4: Market value and time series plot of selected cryptocurrecy and selected time period
