{
"cells": [
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"# Exogenous Variables"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"Exogenous variables can provide additional information to greatly improve forecasting accuracy. Some examples include price or future promotions variables for demand forecasting, and weather data for electricity load forecast. In this notebook we show an example on how to add different types of exogenous variables to NeuralForecast models for making day-ahead hourly electricity price forecasts (EPF) for France and Belgium markets."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"All NeuralForecast models are capable of incorporating exogenous variables to model the following conditional predictive distribution:\n",
"$$\\mathbb{P}(\\mathbf{y}_{t+1:t+H} \\;|\\; \\mathbf{y}_{[:t]},\\; \\mathbf{x}^{(h)}_{[:t]},\\; \\mathbf{x}^{(f)}_{[:t+H]},\\; \\mathbf{x}^{(s)} )$$\n",
"\n",
"where the regressors are static exogenous $\\mathbf{x}^{(s)}$, historic exogenous $\\mathbf{x}^{(h)}_{[:t]}$, exogenous available at the time of the prediction $\\mathbf{x}^{(f)}_{[:t+H]}$ and autorregresive features $\\mathbf{y}_{[:t]}$. Depending on the [train loss](https://nixtla.github.io/neuralforecast/losses.pytorch.html), the model outputs can be point forecasts (location estimators) or uncertainty intervals (quantiles)."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"We will show you how to include exogenous variables in the data, specify variables to a model, and produce forecasts using future exogenous variables."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
":::{.callout-important}\n",
"This Guide assumes basic knowledge on the NeuralForecast library. For a minimal example visit the [Getting Started](./Getting_Started.ipynb) guide.\n",
":::"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"You can run these experiments using GPU with Google Colab.\n",
"\n",
""
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1. Libraries"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"%%capture\n",
"!pip install neuralforecast"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## 2. Load data"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"The `df` dataframe contains the target and exogenous variables past information to train the model. The `unique_id` column identifies the markets, `ds` contains the datestamps, and `y` the electricity price.\n",
"\n",
"Include both historic and future temporal variables as columns. In this example, we are adding the system load (`system_load`) as historic data. For future variables, we include a forecast of how much electricity will be produced (`gen_forecast`) and day of week (`week_day`). Both the electricity system demand and offer impact the price significantly, including these variables to the model greatly improve performance, as we demonstrate in Olivares et al. (2022). \n",
"\n",
"The distinction between historic and future variables will be made later as parameters of the model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize=(15,5))\n",
"plt.plot(df[df['unique_id']=='FR']['ds'], df[df['unique_id']=='FR']['y'])\n",
"plt.xlabel('Date')\n",
"plt.ylabel('Price [EUR/MWh]')\n",
"plt.grid()"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"Add the static variables in a separate `static_df` dataframe. In this example, we are using one-hot encoding of the electricity market. The `static_df` must include one observation (row) for each `unique_id` of the `df` dataframe, with the different statics variables as columns."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
unique_id
\n",
"
market_0
\n",
"
market_1
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
FR
\n",
"
1
\n",
"
0
\n",
"
\n",
"
\n",
"
1
\n",
"
BR
\n",
"
0
\n",
"
1
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" unique_id market_0 market_1\n",
"0 FR 1 0\n",
"1 BR 0 1"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"static_df = pd.read_csv('https://datasets-nixtla.s3.amazonaws.com/EPF_FR_BE_static.csv')\n",
"static_df.head()"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3. Training with exogenous variables"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"We distinguish the exogenous variables by whether they reflect static or time-dependent aspects of the modeled data.\n",
"\n",
"* **Static exogenous variables**: \n",
"The static exogenous variables carry time-invariant information for each time series. When the model is built with global parameters to forecast multiple time series, these variables allow sharing information within groups of time series with similar static variable levels. Examples of static variables include designators such as identifiers of regions, groups of products, etc.\n",
"\n",
"* **Historic exogenous variables**:\n",
"This time-dependent exogenous variable is restricted to past observed values. Its predictive power depends on Granger-causality, as its past values can provide significant information about future values of the target variable $\\mathbf{y}$.\n",
"\n",
"* **Future exogenous variables**: \n",
"In contrast with historic exogenous variables, future values are available at the time of the prediction. Examples include calendar variables, weather forecasts, and known events that can cause large spikes and dips such as scheduled promotions."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"To add exogenous variables to the model, first specify the name of each variable from the previous dataframes to the corresponding model hyperparameter during initialization: `futr_exog_list`, `hist_exog_list`, and `stat_exog_list`. We also set `horizon` as 24 to produce the next day hourly forecasts, and set `input_size` to use the last 5 days of data as input. "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"from neuralforecast.auto import NHITS, BiTCN\n",
"from neuralforecast.core import NeuralForecast\n",
"\n",
"import logging\n",
"logging.getLogger(\"pytorch_lightning\").setLevel(logging.WARNING)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ospra\\miniconda3\\envs\\neuralforecast\\lib\\site-packages\\pytorch_lightning\\utilities\\parsing.py:199: Attribute 'loss' is an instance of `nn.Module` and is already saved during checkpointing. It is recommended to ignore them using `self.save_hyperparameters(ignore=['loss'])`.\n",
"Seed set to 1\n",
"Seed set to 1\n"
]
}
],
"source": [
"horizon = 24 # day-ahead daily forecast\n",
"models = [NHITS(h = horizon,\n",
" input_size = 5*horizon,\n",
" futr_exog_list = ['gen_forecast', 'week_day'], # <- Future exogenous variables\n",
" hist_exog_list = ['system_load'], # <- Historical exogenous variables\n",
" stat_exog_list = ['market_0', 'market_1'], # <- Static exogenous variables\n",
" scaler_type = 'robust'),\n",
" BiTCN(h = horizon,\n",
" input_size = 5*horizon,\n",
" futr_exog_list = ['gen_forecast', 'week_day'], # <- Future exogenous variables\n",
" hist_exog_list = ['system_load'], # <- Historical exogenous variables\n",
" stat_exog_list = ['market_0', 'market_1'], # <- Static exogenous variables\n",
" scaler_type = 'robust',\n",
" ), \n",
" ]"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
":::{.callout-tip}\n",
"When including exogenous variables always use a scaler by setting the `scaler_type` hyperparameter. The scaler will scale all the temporal features: the target variable `y`, historic and future variables.\n",
":::"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
":::{.callout-important}\n",
"Make sure future and historic variables are correctly placed. Defining historic variables as future variables will lead to data leakage.\n",
":::"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"Next, pass the datasets to the `df` and `static_df` inputs of the `fit` method."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"%%capture\n",
"nf = NeuralForecast(models=models, freq='H')\n",
"nf.fit(df=df,\n",
" static_df=static_df)"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4. Forecasting with exogenous variables"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"Before predicting the prices, we need to gather the future exogenous variables for the day we want to forecast. Define a new dataframe (`futr_df`) with the `unique_id`, `ds`, and future exogenous variables. There is no need to add the target variable `y` and historic variables as they won't be used by the model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
unique_id
\n",
"
ds
\n",
"
gen_forecast
\n",
"
week_day
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
FR
\n",
"
2016-11-01 00:00:00
\n",
"
49118.0
\n",
"
1
\n",
"
\n",
"
\n",
"
1
\n",
"
FR
\n",
"
2016-11-01 01:00:00
\n",
"
47890.0
\n",
"
1
\n",
"
\n",
"
\n",
"
2
\n",
"
FR
\n",
"
2016-11-01 02:00:00
\n",
"
47158.0
\n",
"
1
\n",
"
\n",
"
\n",
"
3
\n",
"
FR
\n",
"
2016-11-01 03:00:00
\n",
"
45991.0
\n",
"
1
\n",
"
\n",
"
\n",
"
4
\n",
"
FR
\n",
"
2016-11-01 04:00:00
\n",
"
45378.0
\n",
"
1
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" unique_id ds gen_forecast week_day\n",
"0 FR 2016-11-01 00:00:00 49118.0 1\n",
"1 FR 2016-11-01 01:00:00 47890.0 1\n",
"2 FR 2016-11-01 02:00:00 47158.0 1\n",
"3 FR 2016-11-01 03:00:00 45991.0 1\n",
"4 FR 2016-11-01 04:00:00 45378.0 1"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"futr_df = pd.read_csv('https://datasets-nixtla.s3.amazonaws.com/EPF_FR_BE_futr.csv')\n",
"futr_df['ds'] = pd.to_datetime(futr_df['ds'])\n",
"futr_df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
":::{.callout-important}\n",
"Make sure `futr_df` has informations for the entire forecast horizon. In this example, we are forecasting 24 hours ahead, so `futr_df` must have 24 rows for each time series.\n",
":::"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"Finally, use the `predict` method to forecast the day-ahead prices. "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ospra\\miniconda3\\envs\\neuralforecast\\lib\\site-packages\\utilsforecast\\processing.py:352: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
" freq = pd.tseries.frequencies.to_offset(freq)\n",
"c:\\Users\\ospra\\miniconda3\\envs\\neuralforecast\\lib\\site-packages\\utilsforecast\\processing.py:404: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
" freq = pd.tseries.frequencies.to_offset(freq)\n",
"c:\\Users\\ospra\\OneDrive\\Phd\\Repositories\\neuralforecast\\neuralforecast\\tsdataset.py:91: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
" self.temporal = torch.tensor(temporal, dtype=torch.float)\n",
"c:\\Users\\ospra\\OneDrive\\Phd\\Repositories\\neuralforecast\\neuralforecast\\tsdataset.py:95: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
" self.static = torch.tensor(static, dtype=torch.float)\n",
"c:\\Users\\ospra\\miniconda3\\envs\\neuralforecast\\lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:441: The 'predict_dataloader' does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` to `num_workers=19` in the `DataLoader` to improve performance.\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "35847892c983422d96ad9e8afee27afb",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: | | 0/? [00:00\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
ds
\n",
"
NHITS
\n",
"
BiTCN
\n",
"
\n",
"
\n",
"
unique_id
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
BE
\n",
"
2016-11-01 00:00:00
\n",
"
38.138920
\n",
"
41.105774
\n",
"
\n",
"
\n",
"
BE
\n",
"
2016-11-01 01:00:00
\n",
"
34.647514
\n",
"
35.589905
\n",
"
\n",
"
\n",
"
BE
\n",
"
2016-11-01 02:00:00
\n",
"
33.428795
\n",
"
33.034309
\n",
"
\n",
"
\n",
"
BE
\n",
"
2016-11-01 03:00:00
\n",
"
32.428146
\n",
"
30.183418
\n",
"
\n",
"
\n",
"
BE
\n",
"
2016-11-01 04:00:00
\n",
"
31.068453
\n",
"
29.396011
\n",
"
\n",
" \n",
"
\n",
""
],
"text/plain": [
" ds NHITS BiTCN\n",
"unique_id \n",
"BE 2016-11-01 00:00:00 38.138920 41.105774\n",
"BE 2016-11-01 01:00:00 34.647514 35.589905\n",
"BE 2016-11-01 02:00:00 33.428795 33.034309\n",
"BE 2016-11-01 03:00:00 32.428146 30.183418\n",
"BE 2016-11-01 04:00:00 31.068453 29.396011"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"Y_hat_df = nf.predict(futr_df=futr_df)\n",
"Y_hat_df.head()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"plot_df = df[df['unique_id']=='FR'].tail(24*5).reset_index(drop=True)\n",
"Y_hat_df = Y_hat_df.reset_index(drop=False)\n",
"Y_hat_df = Y_hat_df[Y_hat_df['unique_id']=='FR']\n",
"\n",
"plot_df = pd.concat([plot_df, Y_hat_df ]).set_index('ds') # Concatenate the train and forecast dataframes\n",
"\n",
"plot_df[['y', 'NHITS', 'BiTCN']].plot(linewidth=2)\n",
"plt.axvline('2016-11-01', color='red')\n",
"plt.ylabel('Price [EUR/MWh]', fontsize=12)\n",
"plt.xlabel('Date', fontsize=12)\n",
"plt.grid()"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"In summary, to add exogenous variables to a model make sure to follow the next steps:\n",
"\n",
"1. Add temporal exogenous variables as columns to the main dataframe (`df`).\n",
"2. Add static exogenous variables with the `static_df` dataframe.\n",
"3. Specify the name for each variable in the corresponding model hyperparameter.\n",
"4. If the model uses future exogenous variables, pass the future dataframe (`futr_df`) to the `predict` method."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## References"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"- [Kin G. Olivares, Cristian Challu, Grzegorz Marcjasz, Rafał Weron, Artur Dubrawski, Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx, International Journal of Forecasting](https://www.sciencedirect.com/science/article/pii/S0169207022000413)\n",
"\n",
"- [Cristian Challu, Kin G. Olivares, Boris N. Oreshkin, Federico Garza, Max Mergenthaler-Canseco, Artur Dubrawski (2021). NHITS: Neural Hierarchical Interpolation for Time Series Forecasting. Accepted at AAAI 2023.](https://arxiv.org/abs/2201.12886)"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "python3",
"language": "python",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
![\"Open](\"https://colab.research.google.com/assets/colab-badge.svg\"){kind=icon}