_SeasonalMovingAverageModel

class _SeasonalMovingAverageModel(window: int = 5, seasonality: int = 7)

Bases: object

Seasonal moving average.

$$y_t=\frac{\sum _{i=1}^n{y}_{t-is}}{n},$$

where $s$ is seasonality, $n$ is window size (how many history values are taken for forecast).

Initialize seasonal moving average model.

Length of the context is window * seasonality.

Parameters

• window (int) – Number of values taken for forecast for each point.

• seasonality (int) – Lag between values taken for forecast.

Inherited-members

Methods

fit(df, regressors)	Fit SeasonalMovingAverage model.
forecast(df, prediction_size)	Compute autoregressive forecasts.
predict(df, prediction_size)	Compute predictions using true target data as context.

fit(df: pandas.core.frame.DataFrame, regressors: List[str]) → etna.models.seasonal_ma._SeasonalMovingAverageModel

Fit SeasonalMovingAverage model.

Parameters

• df (pandas.core.frame.DataFrame) – Data to fit on

• regressors (List[str]) – List of the columns with regressors(ignored in this model)

Returns

Fitted model

Return type

etna.models.seasonal_ma._SeasonalMovingAverageModel

forecast(df: pandas.core.frame.DataFrame, prediction_size: int) → numpy.ndarray

Compute autoregressive forecasts.

Parameters

• df (pandas.core.frame.DataFrame) – Features dataframe.

• prediction_size (int) – Number of last timestamps to leave after making prediction. Previous timestamps will be used as a context for models that require it.

Returns

Array with predictions.

Raises

• ValueError: – if context isn’t big enough

• ValueError: – if forecast context contains NaNs

Return type

numpy.ndarray

predict(df: pandas.core.frame.DataFrame, prediction_size: int) → numpy.ndarray

Compute predictions using true target data as context.

Parameters

• df (pandas.core.frame.DataFrame) – Features dataframe.

• prediction_size (int) – Number of last timestamps to leave after making prediction. Previous timestamps will be used as a context for models that require it.

Returns

Array with predictions.

Raises

• ValueError: – if context isn’t big enough

• ValueError: – if there are NaNs in a target column on timestamps that are required to make predictions

Return type

numpy.ndarray