eduzhai > Applied Sciences > Computer Science >

Markov Switching

  • Peter
  • (0) Download
  • 20210219
  • Save

... pages left unread,continue reading

Document pages: 18 pages

Abstract: Markov switching models are a popular family of models that introducestime-variation in the parameters in the form of their state- or regime-specificvalues. Importantly, this time-variation is governed by a discrete-valuedlatent stochastic process with limited memory. More specifically, the currentvalue of the state indicator is determined only by the value of the stateindicator from the previous period, thus the Markov property, and thetransition matrix. The latter characterizes the properties of the Markovprocess by determining with what probability each of the states can be visitednext period, given the state in the current period. This setup decides on thetwo main advantages of the Markov switching models. Namely, the estimation ofthe probability of state occurrences in each of the sample periods by usingfiltering and smoothing methods and the estimation of the state-specificparameters. These two features open the possibility for improvedinterpretations of the parameters associated with specific regimes combinedwith the corresponding regime probabilities, as well as for improvedforecasting performance based on persistent regimes and parameterscharacterizing them.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×