Bayesian analysis of HMM-GARCH models in Finance - Research Proposal Example

The Bayesian approach allows small sample outcomes, fast evaluation, model bigotry and credible reports concerning non-linear roles of the model constraints. Reasonably based financial verdicts hold a gigantic normative characteristic (Bijak, 45). This report paper will explain in detail its purpose, objectives, methodology, limitations and ethical concerns in the study. The first four chapters bring in the research work and an overview of Bayesian analysis of the HMM-GARCH models in Finance.

The subsequent two chapters illustrate the assessment of the HMM-GARCH models with standard improvements. Real financial data is used based on this estimate models. It is noted that still for hefty data analysis the perimeter calculate approximately and distance varies between the two models. Care must be used when basing judgments for these two classes of models. The last two chapters reflect on the limitations and ethical concerns associated with these two models. Introduction to the Problem A particle filtering technique is offered to chronological evaluation that will erect on the change- point model of Chib.

GARCH models can not be estimated with an unidentified quantity of states through subsisting MCMC procedures. No procedures of computing trivial probabilities of these models exist. Therefore, it is highly not convenient to approximate these categories of models by using at hand MCMC methods. This can be possible if one is ready to assume that the integer of break points is also called a priori (Sebe et al, 36). Centre of attention must be on the in order filtering issue other than the smoothing issue of MCMC model.

The path reliance that structural breaks persuade in GARCH models is removed. This is due to the main reason that merely the one-step-ahead prognostic sharing is needed in computation. This therefore, is a fundamental point in excess of two potential states unconfirmed on restrictions in the proposed structural fracture model (Francq et al, 37). Purpose of the Study The Bayesian analysis of HMM-GARCH models in Finance permits the figure of breaks as well as models to be used in this research.

Algorithms made up approximated the model parameters and the integers of structural breaks at each indicate. This is founded on a particular run of the particle filter algorithm. This therefore, makes the models use to be computationally proficiency (Terrell, 27). The confronting global scenery of set models administration is set apart by the ambiguity of the financial markets. The financial sector is always in an invariable activity. A good example of this; is the financial stock market where financial figures change at every moment.

Therefore, the financial trade market is at constant change of financial integers. Incessantly, transforming the jeopardy or profit models manipulate on the latent of variation of intercontinental plus points distribution. The regime-switching models demonstrated enables, the two dissimilar systems. This is by the careful computation of the self-motivated risk or profit structure of any international benefit. The additional room to take account of singular asset types; for example alternative assets, stocks and bonds and in an