SMC Procedure

In risk management, asset pricing, and portfolio optimization, volatility forecasting is critical. The stochastic volatility model, as a special case of nonlinear non-Gaussian state space models, is one of the most popular tools for volatility forecasting. However, there was no closed-form solution for such a sophisticated model until the simulation-based solution, especially the SMC method, became available.

In epidemiology, the SMC method can be used to estimate the stochastic susceptible-infected-recovered (SSIR) model, one of the most popular and accurate models used in epidemiology modeling.

In engineering, the SMC method has been applied in navigation and object tracking.

In Internet of Things (IoT), the SMC method can be used to estimate the states of the machines based on the observable information.

In bioinformatics, the SMC method is used for haplotype inference.

In marketing, the SMC method is used to model the effect of advertising on demand.

three types of SMC filtering algorithms: the sequential importance resampling algorithm, the bootstrap filter algorithm, and the auxiliary particle filter algorithm

two types of SMC smoothing algorithms: the fixed-lag smoothing algorithm and the fixed-interval smoothing algorithm

particle Markov chain Monte Carlo parameter estimation method

proposal distribution for the initial state and transition distributions

simulation and summary estimates of output of the filtering, smoothing, evaluating, and forecasting estimates

output of the log likelihood up to any observation

saving the model information and using it later to score other new data

enables you to run on a cluster of machines that distribute the data and the computations

exploits all the available cores and concurrent threads