Analytical Derivatives for Markov Switching Models
- 69 Downloads
This paper presents analytical gradients for a broad class of regime-switching models with Markovian state-transition probabilities. Such models are usually estimated by maximum likelihood methods, which require the derivatives of the likelihood function with respect to the parameter vector. These gradients are usually calculated by means of numerical techniques. The paper shows that analytical gradients considerably speed up maximum-likelihood estimation.
Unable to display preview. Download preview PDF.
- Diebold, Francis X., Haeng Lee, Joon and Weinbach, Gretchen C. 1994, Regime switching with time varying transition probabilities, in: Hargreaves C. Nonstationary Time Series Analysis and Cointegration, Oxford University Press, Oxford.Google Scholar
- Durland, J. Michael and McCurdy, Thomas H. 1994, Duration dependent transitions in a Markov model of US GNP Growth, Journal of Business & Economic Statistics 12 (July): 279–88.Google Scholar
- Gable, Jeffrey, Norden, Simon van, and Vigfusson, Robert 1995, Analytical derivatives for Markov switching models, Bank of Canada working paper 95-7.Google Scholar
- Goldfeld, S.M. and Quandt, R.E. 1973, A Markov model for switching regressions, Journal of Econometrics 1: 3–16.Google Scholar
- Hamilton, James D. 1989, A new approach to the economic analysis of nonstationary time series and the business cycle, Econometrica 57: 357–384.Google Scholar
- Hamilton, James D. 1992, Estimation, inference, and forecasting of time series subject to changes in regime, in: Rao, C.R. and Maddala, G.S., Handbook of Statistics, Vol. 11, Elsevier Science Publishers B.V.Google Scholar
- Hamilton, James D. 1994, Time Series Analysis, Princeton University Press, Princeton.Google Scholar
- Kim, Chaing-Jin 1994, Dynamic linear models with Markov-switching, Journal of Econometric 60: 1–22.Google Scholar
- van Norden, Simon and Vigfusson, Robert 1996, Regime-switching models; a guide to the Bank of Canada GAUSS procedures, Bank of Canada, working paper 96-3.Google Scholar