ostatslib.actions.regressors.linear_models

Linear regression models actions module

Classes

AutoARIMARegression()	Auto ARIMA regression action.
GammaRegression()	Gamma regression action.
OLSLinearRegression()	Ordinary least squares linear regression action.
PoissonRegression()	Poisson regression action.

class AutoARIMARegression

Auto ARIMA regression action. Fits an ARIMA model based on StatsForecast AutoARIMA model https://nixtla.github.io/statsforecast/src/core/models.html#autoarima-1

Raises:

• ValueError – no time variable assigned

• ValueError – cannot infer frequency

class GammaRegression

Gamma regression action. Fits a Scikit-Learn GammaRegressor https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.GammaRegressor.html

class OLSLinearRegression

Ordinary least squares linear regression action. Fits a Scikit-Learn LinearRegression https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html

execute(data: DataFrame, state: State, config: Config) → tuple[ostatslib.states.state.State, float, ostatslib.actions.base.ActionInfo[sklearn.linear_model._base.LinearRegression]]

Executes action

Parameters:

• data (DataFrame) – data

• state (State) – state

• config (Config) – configuration dictionary

Returns:

tuple of next state, reward and action info

Return type:

tuple[State, float, ActionInfo]

class PoissonRegression

Poisson regression action. Fits a Scikit-Learn PoissonRegressor https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.PoissonRegressor.html