| Z.Q. Lu and L.M. Berliner. "Markov switching time series models with application to a daily runoff series". Water Resour. Res., vol. 35, pp.523-534, 1999. |
....several hundreds and will be reported in greater length in a future publication [20] In this paper a different approach is followed, which is based on HMM s. 3 Hidden Markov Models We now present a HMM formulation of the time series segmentation problem. HMM s have been used for runoff modeling [25] and the possibility of using them for hydrological time series segmentation has been mentioned in [30] but, as far as the author knows, an actual implementation has not been presented yet. On the other hand, we have already mentioned that HMM s are used for segmentation of time series in several ....
Z.Q. Lu and L.M. Berliner. "Markov switching time series models with application to a daily runoff series". Water Resour. Res., vol. 35, pp.523-534, 1999.
.... and has been used in a Bayesian setting in [13] the possibility of using HMM s for hydrological time series segmentation was mentioned earlier in [35] A related approach appears in [42, 43, 44] Other aMM based models appear in [1] 14] 19] 20, 21] and (in combination with AR models) in [29, 45]. According to the previously mentioned association of segments with HMM states, these papers are implicitly connected to multiple change point segmen tation. However, the main goal of these papers is modelling the precipitation process, predicting rain levels etc. Hence their point of view is ....
Z.Q. Lu and L.M. Berliner. "Markov switching time series models with application to a daily runoff series". Water Resour. Res., vol. 35, pp.523-534, 1999.
....as random quantities whose distributions are based on qualitative features of SST evolution. Speci cally, these models will condition on observable quantities that serve as indicators of regime and prognosticators of regime 10 switching (e.g. a wind statistic) Similar models were used in Lu and Berliner (1999) in a di erent context. We consider three regimes corresponding to normal ( normal ) warmer than normal ( warm ) and cooler than normal ( cold ) SST anomalies. Note that, while motivated by ENSO, we do not claim that these regimes are to be interpreted as warm is equivalent to El Ni no or ....
Lu, Z.-Q. and L. M. Berliner, 1999: Markov switching time series models with application to a daily runo series. Water Resour. Res., 35, 523-534.
No context found.
Lu, Z.-Q., and Berliner, L.M. (1999): Markov switching time series models with application to a daily runoff series. Water Resources Research 35, 523-534.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC