_HoltWintersAdapter

class _HoltWintersAdapter(trend: Optional[str] = None, damped_trend: bool = False, seasonal: Optional[str] = None, seasonal_periods: Optional[int] = None, initialization_method: str = 'estimated', initial_level: Optional[float] = None, initial_trend: Optional[float] = None, initial_seasonal: Optional[Sequence[float]] = None, use_boxcox: Union[bool, str, float] = False, bounds: Optional[Dict[str, Tuple[float, float]]] = None, dates: Optional[Sequence[datetime.datetime]] = None, freq: Optional[str] = None, missing: str = 'none', smoothing_level: Optional[float] = None, smoothing_trend: Optional[float] = None, smoothing_seasonal: Optional[float] = None, damping_trend: Optional[float] = None, **fit_kwargs)[source]

Bases: etna.models.base.BaseAdapter

Class for holding Holt-Winters’ exponential smoothing model.

Notes

We use statsmodels.tsa.holtwinters.ExponentialSmoothing model from statsmodels package.

Init Holt-Winters’ model with given params.

Parameters

  • trend (Optional[str]) –

    Type of trend component. One of:

    • ’add’

    • ’mul’

    • ’additive’

    • ’multiplicative’

    • None

  • damped_trend (bool) – Should the trend component be damped.

  • seasonal (Optional[str]) –

    Type of seasonal component. One of:

    • ’add’

    • ’mul’

    • ’additive’

    • ’multiplicative’

    • None

  • seasonal_periods (Optional[int]) – The number of periods in a complete seasonal cycle, e.g., 4 for quarterly data or 7 for daily data with a weekly cycle.

  • initialization_method (str) –

    Method for initialize the recursions. One of:

    • None

    • ’estimated’

    • ’heuristic’

    • ’legacy-heuristic’

    • ’known’

    None defaults to the pre-0.12 behavior where initial values are passed as part of fit. If any of the other values are passed, then the initial values must also be set when constructing the model. If ‘known’ initialization is used, then initial_level must be passed, as well as initial_trend and initial_seasonal if applicable. Default is ‘estimated’. “legacy-heuristic” uses the same values that were used in statsmodels 0.11 and earlier.

  • initial_level (Optional[float]) – The initial level component. Required if estimation method is “known”. If set using either “estimated” or “heuristic” this value is used. This allows one or more of the initial values to be set while deferring to the heuristic for others or estimating the unset parameters.

  • initial_trend (Optional[float]) – The initial trend component. Required if estimation method is “known”. If set using either “estimated” or “heuristic” this value is used. This allows one or more of the initial values to be set while deferring to the heuristic for others or estimating the unset parameters.

  • initial_seasonal (Optional[Sequence[float]]) – The initial seasonal component. An array of length seasonal or length seasonal - 1 (in which case the last initial value is computed to make the average effect zero). Only used if initialization is ‘known’. Required if estimation method is “known”. If set using either “estimated” or “heuristic” this value is used. This allows one or more of the initial values to be set while deferring to the heuristic for others or estimating the unset parameters.

  • use_boxcox ({True, False, 'log', float}, optional) –

    Should the Box-Cox transform be applied to the data first? One of:

    • True

    • False

    • ’log’: apply log

    • float: lambda value

  • bounds (Optional[Dict[str, Tuple[float, float]]]) – An dictionary containing bounds for the parameters in the model, excluding the initial values if estimated. The keys of the dictionary are the variable names, e.g., smoothing_level or initial_slope. The initial seasonal variables are labeled initial_seasonal.<j> for j=0,…,m-1 where m is the number of period in a full season. Use None to indicate a non-binding constraint, e.g., (0, None) constrains a parameter to be non-negative.

  • dates (Optional[Sequence[datetime.datetime]]) – An array-like object of datetime objects. If a Pandas object is given for endog, it is assumed to have a DateIndex.

  • freq (Optional[str]) – The frequency of the time-series. A Pandas offset or ‘B’, ‘D’, ‘W’, ‘M’, ‘A’, or ‘Q’. This is optional if dates are given.

  • missing (str) – Available options are ‘none’, ‘drop’, and ‘raise’. If ‘none’, no nan checking is done. If ‘drop’, any observations with nans are dropped. If ‘raise’, an error is raised. Default is ‘none’.

  • smoothing_level (Optional[float]) – The alpha value of the simple exponential smoothing, if the value is set then this value will be used as the value.

  • smoothing_trend (Optional[float]) – The beta value of the Holt’s trend method, if the value is set then this value will be used as the value.

  • smoothing_seasonal (Optional[float]) – The gamma value of the holt winters seasonal method, if the value is set then this value will be used as the value.

  • damping_trend (Optional[float]) – The phi value of the damped method, if the value is set then this value will be used as the value.

  • fit_kwargs – Additional parameters for calling statsmodels.tsa.holtwinters.ExponentialSmoothing.fit().

Inherited-members

Methods

fit(df, regressors)

Fit Holt-Winters' model.

forecast_components(df)

Estimate forecast components.

get_model()

Get statsmodels.tsa.holtwinters.results.HoltWintersResultsWrapper model that was fitted inside etna class.

predict(df)

Compute predictions from a Holt-Winters' model.

predict_components(df)

Estimate prediction components.
fit(df: pandas.core.frame.DataFrame, regressors: List[str]) → etna.models.holt_winters._HoltWintersAdapter[source]

Fit Holt-Winters’ model.

Parameters

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

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

Returns

Fitted model

Return type

etna.models.holt_winters._HoltWintersAdapter

forecast_components(df: pandas.core.frame.DataFrame) → pandas.core.frame.DataFrame[source]

Estimate forecast components.

Parameters

df (pandas.core.frame.DataFrame) – features dataframe

Returns

dataframe with forecast components

Return type

pandas.core.frame.DataFrame

get_model() → statsmodels.tsa.holtwinters.results.HoltWintersResultsWrapper[source]

Get statsmodels.tsa.holtwinters.results.HoltWintersResultsWrapper model that was fitted inside etna class.

Returns

Internal model

Return type

statsmodels.tsa.holtwinters.results.HoltWintersResultsWrapper

predict(df: pandas.core.frame.DataFrame) → numpy.ndarray[source]

Compute predictions from a Holt-Winters’ model.

Parameters

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

Returns

Array with predictions

Return type

numpy.ndarray

predict_components(df: pandas.core.frame.DataFrame) → pandas.core.frame.DataFrame[source]

Estimate prediction components.

Parameters

df (pandas.core.frame.DataFrame) – features dataframe

Returns

dataframe with prediction components

Return type

pandas.core.frame.DataFrame