Loading...
Please wait, while we are loading the content...
Similar Documents
Nonparametric estimation in Markov switching autoregressive-ARCH models
| Content Provider | Semantic Scholar |
|---|---|
| Author | Kamgaing, Joseph Tadjuidje Stockis, Jean-Pierre Franke, Jürgen |
| Copyright Year | 2010 |
| Abstract | We consider a time series switching between different states which all are characterized as nonparametric AR-ARCH models. The switching is controlled by a hidden Markov chain with finitely many states. We approximate the autoregressive and volatility functions by neural networks and provide an EM algorithm to calculate quasi maximum likelihood estimates of the parameters. A Viterbi algorithm allows to reconstruct the hidden state sequence from the observations. We illustrate the applicability of this approach with a simple portfolio management problem. Finally, we show a consistency result for the network parameters. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.mathematik.uni-kl.de/fileadmin/AGs/stat/Publications/Franke/MSARARCH.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
