I am trying to fit a ARMA(p,q) model using the exact methods where the exact maximum likelihood or conditional sum of squares likelihood is maximized.
This was available in the statsmodels library in version 0.11.1 (https://www.statsmodels.org/v0.11.1/generated/statsmodels.tsa.arima_model.ARMA.fit.html), where the method parameter took values ‘mle’ or ‘css’.
However, in the latest version 0.14.1 (https://www.statsmodels.org/stable/generated/statsmodels.tsa.arima.model.ARIMA.fit.html#statsmodels.tsa.arima.model.ARIMA.fit), the method parameter takes values ‘statespace’, ‘innovations_mle’, ‘hannan_rissanen’, ‘burg’, ‘innovations’, and ‘yule_walker’.
Is there a mapping from {‘mle’,‘css’} to {‘statespace’, ‘innovations_mle’, ‘hannan_rissanen’, ‘burg’, ‘innovations’, ‘yule_walker’}? If not, is there a way to fit the ARMA(p,q) model using Python statsmodels where the exact maximum likelihood or conditional sum of squares likelihood is maximized?