python - Pykalman with non-square observation matrix

Question

In the documentation of Pykalman it says that it only accepts square matrices for the observation_matrices and transition_matrices arguments.

Is there a way around this? I have to estimate a state-space system with a non-square observation matrix. Using the Pykalman notation, its dimensions would be [n_dim_obs, n_dim_state].

I could not find any other kalman filter library that accepts non-square observation matrices and has a loglikelihood method.

Answer 1

我对 Pykalman 有一些经验，它的文档似乎有很多错误。所以观察矩阵不必是方阵。它只需要将估计状态映射到测量矩阵。

请在此处查看我的答案。观察矩阵看起来像 H = [0 0 1] 并且 Pykalman 工作正常。维度正是您所期望的：[n_dim_obs, n_dim_state]。

如果您需要其他信息，我很乐意提供帮助。