Dynamic Logistic State Space Prediction Model
Implements the dynamic logistic state space model for binary outcome data proposed by Jiang et al. (2021) <doi:10.1111/biom.13593>. It provides a computationally efficient way to update the prediction whenever new data becomes available. It allows for both time-varying and time-invariant coefficients, and use cubic smoothing splines to model varying coefficients. The smoothing parameters are objectively chosen by maximum likelihood. The model is updated using batch data accumulated at pre-specified time intervals.