Tracking public opinion over time with Dirichlet

[jessegrabowski]

@ricardoV94

You could parametrize the simplex differently, such as using multiple gaussian random walk and then taking the softmax at each time step (you need one less random walk than the number of dimensions)

The alpha parameter of the Dirichlet shouldn’t itself be a simplex, so one only needs a strictly positive random walk. There’s also the AR distribution if you don’t like the random walk assumption.

If you have features you want to include in the hidden states (e.g. ARIMAX hidden state) you have to do a scan.