For the sake of space and your time, this chapter introduced and applied three filtering and smoothing classes of algorithms. Moving averages, Fourier series, and the Kalman filter are far from being the only techniques used in cleaning raw data. The alternative techniques can be classified into the following categories:
Autoregressive models that encompass Autoregressive Moving Average (ARMA), Autoregressive Integrated Moving Average (ARIMA), generalized autoregressive conditional heteroskedasticity (GARCH), and Box-Jenkins that relies on some form of autocorrelation function.
Curve-fitting algorithms that include the polynomial and geometric fit with the ordinary least squares method, nonlinear least squares using the Levenberg-Marquardt optimizer, and probability distribution fitting.
Nonlinear dynamic systems with a Gaussian noise such as a particle filter.
Hidden Markov models, as described in the The hidden Markov model section in Chapter 7, Sequential...