1: \begin{abstract}

2: Extended dynamic mode decomposition (EDMD) is a data-driven algorithm

3: for approximating spectral data of the

4: Koopman operator associated to a dynamical system,

5: combining a Galerkin method of order $N$ and collocation method of order $M$.

6: Spectral convergence of this method subtly depends on appropriate choice of

7: the space of observables. For

8: chaotic analytic full branch maps of the interval, we derive a constraint between $M$ and $N$ guaranteeing spectral convergence of EDMD. 

9: \end{abstract}

10: