statsmodels.tsa.forecasting.theta.ThetaModelResults

class statsmodels.tsa.forecasting.theta.ThetaModelResults(b0, alpha, sigma2, one_step, seasonal, use_mle, model)[source]

Results class from estimated Theta Models.

Parameters:
b0float

The estimated trend slope.

alphafloat

The estimated SES parameter.

sigma2float

The estimated residual variance from the SES/IMA model.

one_stepfloat

The one-step forecast from the SES.

seasonalndarray

An array of estimated seasonal terms.

use_mlebool

A flag indicating that the parameters were estimated using MLE.

modelThetaModel

The model used to produce the results.

Attributes:
model

The model used to produce the results

params

The forecasting model parameters

sigma2

The estimated residual variance

Methods

forecast([steps, theta])

Forecast the model for a given theta

forecast_components([steps])

Compute the three components of the Theta model forecast

plot_predict([steps, theta, alpha, ...])

Plot forecasts, prediction intervals and in-sample values

prediction_intervals([steps, theta, alpha])

Parameters:

summary()

Summarize the model

Properties

model

The model used to produce the results

params

The forecasting model parameters

sigma2

The estimated residual variance


Last update: Oct 12, 2024