Advanced Sarimax Usage

If you are not a magician who can easily infer correct SARIMAX orders from looking on PACF (partial autocorrelation function) and ACF (autocorrelation function), you want to rather leverage AutoSarima which finds them for you - set init_with_autoarima to True.

If you want to further configure the search space of AutoARIMA, then you can provide all parameters of pmdarima.arima.AutoARIMA as autoarima_dict arguments.

When you run cross-validation with enabled AutoARIMA (init_with_autoarima), it’s often advisable to find the correct order only during the first fit call and reuse this model on all other splits in order to simulate the out-of-sample performance.

The signature of SarimaxWrapper contains parameters of pmdarima.arima.ARIMA, not AutoARIMA class.

For more parameters check pmdarima docs

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
plt.style.use('seaborn')
plt.rcParams['figure.figsize'] = [12, 6]

from hcrystalball.utils import get_sales_data

df = get_sales_data(n_dates=100,
                    n_assortments=1,
                    n_states=1,
                    n_stores=1)
X, y = pd.DataFrame(index=df.index), df['Sales']

from hcrystalball.wrappers import SarimaxWrapper

SarimaxWrapper?

Init signature:
SarimaxWrapper(
    order=None,
    seasonal_order=(0, 0, 0, 0),
    start_params=None,
    method='lbfgs',
    maxiter=50,
    suppress_warnings=False,
    out_of_sample_size=0,
    scoring='mse',
    scoring_args=None,
    trend=None,
    with_intercept=True,
    name='sarimax',
    conf_int=False,
    init_with_autoarima=False,
    autoarima_dict=None,
    always_search_model=False,
    clip_predictions_lower=None,
    clip_predictions_upper=None,
)
Docstring:
Wrapper for `pmdarima.arima.ARIMA` and `pmdarima.arima.AutoARIMA`

Search for optimal order of SARIMAX type model or instantiate one
in case you provide specific order.

Parameters
----------
name : str
    Name of the model instance, used also as column name for returned prediction.

conf_int: bool
    Whether confidence intervals should be also outputed.

init_with_autoarima: bool
    Whether you want to leverage automated search of pmdarima.arima.AutoARIMA.

autoarima_dict: dict
    If `init_with_autoarima` is set to True, then `autoarima_dict` is used for instantiation
    of pmdarima.arima.AutoARIMA class, thus it serves as configuration of AutoARIMA search.

always_search_model: bool
    If `init_with_autoarima` is set to True and `always_search_model` is set to True, then
    the optimal model will be searched for during each `fit` call. On the other hand in most
    cases the desired behaviour is to search for optimal model just for first `fit` call and
    reused this already found model on subsequent `fit` calls (i.e. during cross validation).

clip_predictions_lower: float
    Minimal value allowed for predictions - predictions will be clipped to this value.

clip_predictions_upper: float
    Maximum value allowed for predictions - predictions will be clipped to this value.
File:           ~/data_science/time-series-framework/src/hcrystalball/wrappers/_sarimax.py
Type:           ABCMeta
Subclasses:

model = SarimaxWrapper(
    autoarima_dict={'d':1, 'm':7, 'max_p':2, 'max_q':2},
    init_with_autoarima=True
)

(model.fit(X[:-10], y[:-10])
      .predict(X[-10:])
      .merge(y, left_index=True, right_index=True, how='outer')
      .tail(50)
      .plot()
);

And now access the models parameters

model

SarimaxWrapper(always_search_model=False,
               autoarima_dict={'d': 1, 'm': 7, 'max_p': 2, 'max_q': 2},
               clip_predictions_lower=None, clip_predictions_upper=None,
               conf_int=False, init_with_autoarima=False, maxiter=50,
               method='lbfgs', name='sarimax', order=(2, 1, 2),
               out_of_sample_size=0, scoring='mse', scoring_args=None,
               seasonal_order=(2, 0, 1, 7), start_params=None,
               suppress_warnings=False, trend=None, with_intercept=True)

You might also directly pass the orders if you know, what are you doing

model = SarimaxWrapper(order=(1, 1, 2), seasonal_order=(1, 0, 2, 7))

(model.fit(X[:-10], y[:-10])
      .predict(X[-10:])
      .merge(y, left_index=True, right_index=True, how='outer')
      .tail(50)
      .plot()
);

model

SarimaxWrapper(always_search_model=False, autoarima_dict=None,
               clip_predictions_lower=None, clip_predictions_upper=None,
               conf_int=False, init_with_autoarima=False, maxiter=50,
               method='lbfgs', name='sarimax', order=(1, 1, 2),
               out_of_sample_size=0, scoring='mse', scoring_args=None,
               seasonal_order=(1, 0, 2, 7), start_params=None,
               suppress_warnings=False, trend=None, with_intercept=True)