Created
September 3, 2019 17:42
-
-
Save lamres/900f1e6cf44ab6040f5509d6bcacd50f to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Normalized st. deviation | |
| def std_normalized(vals): | |
| return np.std(vals) / np.mean(vals) | |
| # Ratio of diff between last price and mean value to last price | |
| def ma_ratio(vals): | |
| return (vals[-1] - np.mean(vals)) / vals[-1] | |
| # z-score for volumes and price | |
| def values_deviation(vals): | |
| return (vals[-1] - np.mean(vals)) / np.std(vals) | |
| # Feature params | |
| future_period = 1 | |
| std_period = 10 | |
| ma_period = 10 | |
| price_deviation_period = 10 | |
| volume_deviation_period = 10 | |
| # Create features | |
| cols_features = ['last_return', 'std_normalized', 'ma_ratio', 'price_deviation', 'volume_deviation'] | |
| dataset['last_return'] = dataset[column_price].pct_change() | |
| dataset['std_normalized'] = dataset[column_price].rolling(std_period).apply(std_normalized) | |
| dataset['ma_ratio'] = dataset[column_price].rolling(ma_period).apply(ma_ratio) | |
| dataset['price_deviation'] = dataset[column_price].rolling(price_deviation_period).apply(values_deviation) | |
| dataset['volume_deviation'] = dataset[column_volume].rolling(volume_deviation_period).apply(values_deviation) | |
| dataset["future_return"] = dataset[column_price].pct_change(future_period).shift(-future_period) | |
| dataset = dataset.replace([np.inf, -np.inf], np.nan) | |
| dataset = dataset.dropna() | |
| # Split the data on sets | |
| train_ind = int(np.where(dataset.index == '2018-01-01 00:00:00')[0]) | |
| train_set = dataset[cols_features].values[:train_ind] | |
| test_set = dataset[cols_features].values[train_ind:] | |
| # Plot features | |
| plt.figure(figsize=(20,10)) | |
| fig, axs = plt.subplots(len(cols_features), 1, figsize = (15, 15)) | |
| colours = cm.rainbow(np.linspace(0, 1, len(cols_features))) | |
| for i in range(0, len(cols_features)): | |
| axs[i].plot(dataset.reset_index()[cols_features[i]], color = colours[i]) | |
| axs[i].set_title(cols_features[i]) | |
| axs[i].grid(True) | |
| plt.tight_layout() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment