Hidden Markov Models for Time Series

Zucchini, Walter; MacDonald, Iain L.

doi:10.1201/9781420010893

Cited by 368 publications

(102 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…† Loglikelihood for this model is 1318.23, which is higher than 1212.1 and 1065.47 for a model with two hidden states and a simple GBM, respectively. Following Zucchini and MacDonald (2009), we use two measures of 'lack of fit' of the model: the Akaike information criterion (AIC) and Bayesian information criterion (BIC). AIC has the value of −2608.45, as opposed to −2410.2 and −2126.95, respectively.…”

Section: Estimation Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Estimating animal spirits: conservative risk calculation

Andruszkiewicz

Davis

Lleo

2014

Quantitative Finance Letters

View full text Add to dashboard Cite

In this paper, we estimate behavioural factors-Keynes' 'animal spirits'-in the property market. An enhanced hidden Markov model is used, for both the Shiller Home Price Index and a consumer confidence index. We conclude that both house prices and consumer confidence are driven by another hidden behavioural factor, interpreted as 'animal spirits'. Both data series imply similar paths of the hidden factor. The estimated model is used for value-at-risk calculation and forecasting. We introduce an intuitive method to include crisis scenarios in the model, which proves to produce much better risk estimates during the credit crunch, without substantially affecting the distribution during growth periods.

show abstract

Section: Estimation Resultsmentioning

confidence: 99%

“…See also Zucchini and MacDonald (2009) for a comprehensive treatment of the subject. We extend the standard algorithm to cater for mean-reverting processes, show how to estimate parameters of correlated multidimensional continuous-time stochastic processes and provide a procedure for forecasting.…”

Section: Introductionmentioning

confidence: 99%

Estimating animal spirits: conservative risk calculation

Andruszkiewicz

Davis

Lleo

2014

Quantitative Finance Letters

View full text Add to dashboard Cite

show abstract

“…As already pointed out in Sect. 1, a parameter-driven alternative would be a (two-state) binomial HMM (Zucchini and MacDonald 2009). Although the estimates for the HMM's transition matrix are not significant, there are well-interpretable analogies between the fitted HMM and the LSET-BAR(1) model.…”

Section: Data Example: Measles In Germany In 2004-2005mentioning

confidence: 99%

“…One such approach is hidden Markov models (HMM) for counts (Zucchini and MacDonald 2009), where a state dependence of the observed counts is introduced through an underlying (invisible) finite Markov chain (also see Sect. 5 below).…”

Section: Introductionmentioning

confidence: 99%

Self-exciting threshold binomial autoregressive processes

et al. 2015

View full text Add to dashboard Cite

We introduce a new class of integer-valued self-exciting threshold models, which is based on the binomial autoregressive model of order one as introduced by McKenzie (Water Resour Bull 21:645-650, 1985. doi:10.1111/j.1752-1688.1985. Basic probabilistic and statistical properties of this class of models are discussed. Moreover, parameter estimation and forecasting are addressed. Finally, the performance of these models is illustrated through a simulation study and an empirical application to a set of measle cases in Germany.

show abstract

“…HMMs have already been used to analyze ground-observed wind speed and direction without forecast verification purposes. Most of the time independence between speed and direction is assumed, as in Holzmann et al (2006), Zucchini and MacDonald (2009), Bulla et al (2012) and Bulla et al (2015), but exceptions exist. For example Lagona et al (2015) use the recently introduced Abe-Ley cylindrical density (Abe and Ley, 2015) and Mastrantonio et al (2015a) adopt the general circular-linear projected normal.…”

Section: Introductionmentioning

confidence: 99%

Distributions-oriented wind forecast verification by a hidden Markov model for multivariate circular–linear data

Mastrantonio

Pollice

Fedele

2017

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Winds from the North-West quadrant and lack of precipitation are known to lead to an increase of PM10 concentrations over a residential neighborhood in the city of Taranto (Italy). In 2012 the local government prescribed a reduction of industrial emissions by 10% every time such meteorological conditions are forecasted 72 hours in advance. Wind forecasting is addressed using the Weather Research and Forecasting (WRF) atmospheric simulation system by the Regional Environmental Protection Agency. In the context of distributions-oriented forecast verification, we propose a comprehensive model-based inferential approach to investigate the ability of the WRF system to forecast the local wind speed and direction allowing different performances for unknown weather regimes. Ground-observed and WRF-forecasted wind speed and direction at a relevant location are jointly modeled as a 4-dimensional time series with an unknown finite number of states characterized by homogeneous distributional behavior. The proposed model relies on a mixture of joint projected and skew normal distributions with time-dependent states, where the temporal evolution of the state membership follows a first order Markov process. Parameter estimates, including the number of states, are obtained by a Bayesian MCMC-based method. Results provide useful insights on the performance of WRF forecasts in relation to different combinations of wind speed and direction.

show abstract

Hidden Markov Models for Time Series

Cited by 368 publications

References 0 publications

Estimating animal spirits: conservative risk calculation

Estimating animal spirits: conservative risk calculation

Self-exciting threshold binomial autoregressive processes

Distributions-oriented wind forecast verification by a hidden Markov model for multivariate circular–linear data

Contact Info

Product

Resources

About