Via Hyndsight :

Mathematics is not just solving for X. It’s also figuring out Y (Why). - Arthur Benjamin

Takeaway : The strength of this book is the focus on the simplest state space model and then showing all the aspects of Kalman Filter framework and related pseudo-codes for filtering, smoothing. The novelty of this book is the central focus on the idea of orthogonalization of observation data. This makes all the Kalman Filter related formulae take convenient forms that one can intuitively as well as rigorously understand. One thing missing from this book is the discussion of numerical stability of filtering and smoothing algorithms, considering that the purpose of the book is to enable the reader code up his/her own KF functions.

[youtube https://www.youtube.com/watch?v=LPvtpXtv0Lg?rel=0]

State Space Methodology serves as an umbrella for representing many univariate, multivariate stationary and non stationary time series. For those who have never heard of a “State Space Model” but have used some software for any time series model parameter estimation, the motivation is this : It is likely that the software used has a state space representation of the model in the implementation. For example, in R, the implementation for the function arima() says,