Build sales forecast and user interest predictive models using Prophet.
https://nbviewer.org/github/juil/ProphetForecast/blob/master/forecasting_net_prophet.ipynb
| Period | Google Search Traffic |
|---|---|
| 2016-2020 | 35172.5 |
| May 2020 | 38181 |
| close | Search Trends | Lagged Search Trends | Stock Volatility | Hourly Stock Return | |
|---|---|---|---|---|---|
| close | 1.000000 | 0.011918 | 0.012135 | 0.477935 | 0.022970 |
| Search Trends | 0.011918 | 1.000000 | 0.384292 | -0.180868 | -0.029732 |
| Lagged Search Trends | 0.012135 | 0.384292 | 1.000000 | -0.119010 | 0.018147 |
| Stock Volatility | 0.477935 | -0.180868 | -0.119010 | 1.000000 | 0.046713 |
| Hourly Stock Return | 0.022970 | -0.029732 | 0.018147 | 0.046713 | 1.000000 |
Rendered in Google Colab
# Install the required libraries
!pip install prophet
!pip install hvplot
!pip install holoviews
!pip install tabulate
!pip install numpy*Starter code provided by edX.