No label defined (Q3160503)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 2133112
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 2133112

      Statements

      0 references
      9 February 2005
      0 references
      hidden Markov model
      0 references
      Bayesian
      0 references
      model select
      0 references
      recursive algorithm
      0 references
      The author considers the following dynamic model of the price \(X(t)\) of a single stock at time \(t\), \[ dX(t)=X(t)\mu_{J(t)}dt+ X(t)\sigma_{J(t)}dW(t), \] where \(W(t)\) is a Wiener process, and \(J(t)\) is a Markov process taking discrete values in \(\{1,\dots,S\}\) independently of \(W(t)\). For each state \(j\) there is a drift parameter \(\mu_j\) and a volatility parameter \(\sigma_j\). The above model is referred to by several names, such as the ``Regime/Markov switching model'', the ``Markov modulated geometric Brownian motion model'', etc. In the first part of the paper the author proposes an estimation methodology to fit the model and model selection strategies to determine the optimum number of states within a Bayesian framework. A case study on AT\&T stock price data is reported. Based on the empirical study, a notion of ``regime shift detection'' in financial time series data and a detection method based on the estimation algorithm are developed. In the second part of the paper the author deals with the rate of convergence and error bounds for the Wonham filter with random parameters. The basic idea is to construct approximate filters when only noisy/simulated values are available. These suboptimal filters are proved to converge to the desired Wonham filter under simple ergodicity conditions. For the entire collection see [Zbl 1048.91001].
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references