mne.time_frequency.fit_iir_model_raw

mne.time_frequency.fit_iir_model_raw(raw, order=2, picks=None, tmin=None, tmax=None, verbose=None)

Fits an AR model to raw data and creates the corresponding IIR filter

The computed filter is the average filter for all the picked channels. The frequency response is given by:

\[H(e^{jw}) = \frac{1}{a[0] + a[1]e^{-jw} + ... + a[n]e^{-jnw}}\]
Parameters:

raw : Raw object

an instance of Raw.

order : int

order of the FIR filter.

picks : array-like of int | None

indices of selected channels. If None, MEG and EEG channels are used.

tmin : float

The beginning of time interval in seconds.

tmax : float

The end of time interval in seconds.

verbose : bool, str, int, or None

If not None, override default verbose level (see mne.verbose).

Returns:

b : ndarray

Numerator filter coefficients.

a : ndarray

Denominator filter coefficients