Walter Zucchini

Georg-August-Universität Göttingen, Göttingen, Lower Saxony, Germany

Are you Walter Zucchini?

Claim your profile

Publications (43)28.81 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Stochastic daily precipitation models are commonly used to generate scenarios of climate variability or change on a daily time scale. The standard models consist of two components describing the occurrence and intensity series, respectively. Binary logistic regression is used to fit the occurrence data, and the intensity series is modeled by a continuous-valued right-skewed distribution, such as gamma, Weibull or lognormal. The precipitation series is then modeled using the joint density and standard software for generalized linear models can be used to perform the computations. A drawback of these precipitation models is that they do not produce a sufficiently heavy upper tail for the distribution of daily precipitation amounts; they tend to underestimate the frequency of large storms. In this study we adapted the approach of Furrer and Katz (2008) based on hybrid distributions in order to correct for this shortcoming. In particular we applied hybrid gamma - generalized Pareto (GP) and hybrid Weibull-GP distributions to develop a stochastic precipitation model for daily rainfall at Ihtiman in western Bulgaria. We report the results of simulations designed to compare the models based on the hybrid distributions and those based on the standard distributions. Some potential difficulties are outlined.
    01/2014; 2(2).
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We develop estimators for line transect surveys of animals that are stochastically unavailable for detection while within detection range. The detection process is formulated as a hidden Markov model with a binary state-dependent observation model that depends on both perpendicular and forward distances. This provides a parametric method of dealing with availability bias when estimates of availability process parameters are available even if series of availability events themselves are not. We apply the estimators to an aerial and a shipboard survey of whales, and investigate their properties by simulation. They are shown to be more general and more flexible than existing estimators based on parametric models of the availability process. We also find that methods using availability correction factors can be very biased when surveys are not close to being instantaneous, as can estimators that assume temporal independence in availability when there is temporal dependence.
    Biometrics 07/2013; · 1.41 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We introduce a number of nonstandard stochastic volatility (SV) models and examine their performance when applied to the series of daily returns on several stocks listed on the New York Stock Exchange. The nonstandard models under investigation extend both the observation process and the volatility-generating process of basic SV models. In particular, we consider dependent as well as independent mixtures of autoregressive components as the log-volatility process, and include in the observation equation a lower bound on the volatility. We also consider an experimental SV model that is based on conditionally gamma-distributed volatilities.Our estimation method is based on the fact that an SV model can be approximated arbitrarily accurately by a hidden Markov model (HMM), whose likelihood is easy to compute and to maximize. The method is close, but not identical, to those of Fridman and Harris (1998), Bartolucci and De Luca (2001, 2003) and Clements et al. (2006), and makes explicit the useful link between HMMs and the methods of those authors. Likelihood-based estimation of the parameters of SV models is usually regarded as challenging because the likelihood is a high-dimensional multiple integral. The HMM approximation is easy to implement and particularly convenient for fitting experimental extensions and variants of SV models such as those we introduce here. In addition, and in contrast to the case of SV models themselves, simple formulae are available for the forecast distributions of HMMs, for computing appropriately defined residuals, and for decoding, i.e. estimating the volatility of the process.
    Journal of Empirical Finance - J EMPIR FINANC. 01/2012;
  • Source
    Katja Landau, Stephan Klasen, Walter Zucchini
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigate the accuracy of ex ante assessments of vulnerability to income poverty using cross-sectional data and panel data. We use long-term panel data from Germany and apply different regression models, based on household covariates and previous-year equivalence income, to classify a household as vulnerable or not. Predictive performance is assessed using the Receiver Operating Characteristics (ROC), which takes account of false positive as well as true positive rates. Estimates based on cross-sectional data are much less accurate than those based on panel data, but for Germany, the accuracy of vulnerability predictions is limited even when panel data are used. In part this low accuracy is due to low poverty incidence and high mobility in and out of poverty.
    01/2012;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Many animals are organized into social groups and have to synchronize their activities to maintain group cohesion. Although activity budgets, habitat constraints, and group properties may impact on behavioural synchrony, little is known regarding how members of a group reach a consensus on the timing of activities such as foraging bouts. Game theory predicts that pair partners should synchronize their activities when there is an advantage of foraging together. As a result of this synchronization, differences in the energetic reserves of the two players develop spontaneously and the individual with lower reserves emerges as a pacemaker of the synchrony. Here, we studied the behavioral synchrony of pair-living, nocturnal, red-tailed sportive lemurs (Lepilemur ruficaudatus). We observed 8 pairs continuously for ≥1 annual reproductive cycle in Kirindy Forest, Western Madagascar. During focal observations, one observer followed the female of a pair and, simultaneously, another observer followed the male. We recorded the location and behavioral state of the focal individual every 5 min via instantaneous sampling. Although behavioral synchrony of pair partners appeared to be due mainly to endogenous activity patterns, they actively synchronized when they were in visual contact (<10 m). Nevertheless, red-tailed sportive lemurs benefit from synchronizing their activity only for 15% of the time, when they are close together. The lack of an early warning system for predators and weak support for benefits via social information transfer in combination with energetic constraints may explain why red-tailed sportive lemurs do not spend more time together and thus reap the benefits of behavioral synchrony.
    International Journal of Primatology 12/2011; 32(6):1383-1396. · 1.79 Impact Factor
  • Source
    Jing Dai, Stefan Sperlich, Walter Zucchini
    [Show abstract] [Hide abstract]
    ABSTRACT: A reliable prediction of unconditional welfare distributions, like income or consumption, is essential for welfare analysis, and in particular for inequality, poverty or development studies. Where observations of expenditures or income are missing, the mean prediction based on available covariates is not just a poor estimator of the unconditional distribution; it fails to predict the required information about tails and quantiles. A new estimation method is introduced which can be combined with any mean prediction model. It is used to calculate the income distribution of a survey based on subsample information, to estimate the unconditional income distribution for the non-responding households, and to predict the household expenditures of a future panel wave. It allows for imputing welfare distributions for a census from a survey or for synthetic populations under specific scenarios. Further inference is straight-forward, including prediction of Lorenz curves, indexes like the Gini, or distribution quantiles, including confidence intervals.
    11/2011;
  • Source
    Jing Dai, Walter Zucchini, Stefan Sperlich
    [Show abstract] [Hide abstract]
    ABSTRACT: In many countries the demand for health care services is of increasing importance. Especially in the industrialized world with a changing demographic structure social insurances and politics face real challenges. Reliable predictors of those demand functions will therefore become invaluable tools. This article proposes a prediction method for the distribution of the number of visits to the medical doctor for a determined population, given a sample that is not necessarily taken from that population. It uses the estimated conditional sample distribution, and it can be applied for forecast scenarios. The methods are illustrated along data from Sidney. The introduced methodology can be applied as well to any other prediction problem of discrete distributions in real, future or any fictitious population. It is therefore also an excellent tool for future predictions, scenarios and policy evaluation.
    11/2011;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Adult-onset Still's disease (AOSD) is a potentially crippling or life-threatening rare disease that may be self-limited, intermittent, and chronic. Clinical predictors of outcome are still lacking, as is information on the rate of progress of its chronic course. The main objective is to identify factors that improve our ability to predict the course of AOSD, and factors associated with the rate of progress of its chronic course. A comparison with the literature is included. A retrospective cohort observational study conducted at the tertiary-referral Rheumatology Unit in Ferrara, Italy. Seventy-six patients (44 females and 32 males) referred to the Unit and who satisfied the criteria for AOSD were identified. Our findings on white AOSD patients are largely compatible with those previously published. Ferritin level, as well disease activity score (DAS(28)), is associated with the rate of progression of the articular manifestations of the disease. A polyarthritis persisting over 6 months is associated with the development of a chronic articular course, irrespective of the size of the involved joints. Ferritin, being associated with the course of AOSD, could play a role in the diagnosis of the disease. Together with DAS(28), it might also serve as a useful predictor for the rate of progress of the chronic course of the disease, as measured with simple erosion narrowing score.
    Seminars in arthritis and rheumatism 03/2011; 41(2):279-85. · 4.72 Impact Factor
  • R. Langrock, W. Zucchini
    [Show abstract] [Hide abstract]
    ABSTRACT: A hidden Markov model (HMM) with a special structure that captures the ‘semi’-property of hidden semi-Markov models (HSMMs) is considered. The proposed model allows arbitrary dwell-time distributions in the states of the Markov chain. For dwell-time distributions with finite support the HMM formulation is exact while for those that have infinite support, e.g. the Poisson, the distribution can be approximated with arbitrary accuracy. A benefit of using the HMM formulation is that it is easy to incorporate covariates, trend and seasonal variation particularly in the hidden component of the model. In addition, the formulae and methods for forecasting, state prediction, decoding and model checking that exist for ordinary HMMs are applicable to the proposed class of models. An HMM with explicitly modeled dwell-time distributions involving seasonality is used to model daily rainfall occurrence for sites in Bulgaria.
    Computational Statistics & Data Analysis 01/2011; · 1.30 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The Human Development Index (HDI) published in the Human Development Report (HDR) of the United Nations Development Program is calculated as a simple average of the Life Expectancy Index (LEI), the Education Index (EI) and the Gross Domestic Product Index (GDPI). This paper provides statistical support for the use of this seemingly arbitrary equal weighting of the three components by treating human development as a latent concept imperfectly captured by its three component indices. We show that a principal component analysis (PCA) based on the correlation matrix of the components leads to practically the same weights. Specifically we show that, for the period 1975 to 2005, the first principal component accounts for between 78% and 90% of the total variability in the data, and that its coefficients are positive and nearly equal. By normalizing the coefficients, the simple average weighting (1/3, 1/3, 1/3) scheme is obtained. The ranks of countries obtained using the PCA weightings are very similar to those based on the HDI. An advantage of the simple equal weighting is that one can define a simple index to measure the balance of a country\'s development, given its HDI which we show below.
    Courant Research Centre PEG, Courant Research Centre: Poverty, Equity and Growth - Discussion Papers. 01/2010;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Tick-borne encephalitis (TBE) virus can cause severe symptoms in humans. The incidence of this vector-borne pathogen in humans is characterised by spatial and temporal heterogeneity. To explain the variation in reported human TBE cases per county in southern Germany, we designed a time-lagged, spatially-explicit model that incorporates ecological, environmental, and climatic factors. We fitted a logistic regression model to the annual counts of reported human TBE cases in each of 140 counties over an eight year period. The model controlled for spatial autocorrelation and unexplained temporal variation. The occurrence of human TBE was found to be positively correlated with the proportions of broad-leafed, mixed and coniferous forest cover. An index of forest fragmentation was negatively correlated with TBE incidence, suggesting that infection risk is higher in fragmented landscapes. The results contradict previous evidence regarding the relevance of a specific spring-time temperature regime for TBE epidemiology. Hunting bag data of roe deer (Capreolus capreolus) in the previous year was positively correlated with human TBE incidence, and hunting bag density of red fox (Vulpes vulpes) and red deer (Cervus elaphus) in the previous year were negatively correlated with human TBE incidence. Our approach suggests that a combination of landscape and climatic variables as well as host-species dynamics influence TBE infection risk in humans. The model was unable to explain some of the temporal variation, specifically the high counts in 2005 and 2006. Factors such as the exposure of humans to infected ticks and forest rodent population dynamics, for which we have no data, are likely to be explanatory factors. Such information is required to identify the determinants of TBE more reliably. Having records of TBE infection sites at a finer scale would also be necessary.
    International Journal of Health Geographics 01/2010; 9:42. · 2.62 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Jeder hat eine Vorstellung davon, was man unter Statistik versteht, und viele denken dabei sicherlich zunächst an umfangreiche Tabellen oder grafische Darstellungen, die bestimmte Sachverhalte in komprimierter Weise verdeutlichen. Dies ist jedoch nur ein Teil der Statistik, die sogenannte beschreibende oder deskriptive Statistik, die dazu dient, umfangreiche Datensätze mit Hilfe von Abbildungen und Kennzahlen anschaulich darzustellen. In den meisten Fällen geht die Statistik jedoch weit über die reine Beschreibung von Datensätzen hinaus. In der Regel sind vorliegende Daten nur eine Stichprobe aus einer sogenannten Grundgesamtheit, und man möchte aus der Stichprobe Schlussfolgerungen für die Grundgesamtheit ziehen. Dieser Teil der Statistik wird schließende oder induktive Statistik genannt. Zu Beginn dieses ersten Kapitels werden zunächst einige praktische Anwendungsbeispiele statistischer Methoden vorgestellt, um einen Eindruck von den vielfältigen Anwendungsmöglichkeiten der Statistik zu vermitteln. Im hinteren Teil des Kapitels werden dann einige wichtige Grundbegriffe der Statistik, wie zum Beispiel Stichprobe und Grundgesamtheit, eingeführt.
    04/2009: pages 1-40;
  • Statistik für Bachelor- und Masterstudenten: Eine Einführung für Wirtschafts- und Sozialwissenschaftler, Statistik und ihre Anwendungen. ISBN 978-3-540-88986-1. Springer Berlin Heidelberg, 2009. 01/2009;
  • [Show abstract] [Hide abstract]
    ABSTRACT: We describe a family of models developed for time series of animal feeding behavior. The models incorporate both an unobserved state, which can be interpreted as the motivational state of the animal, and a mechanism for feedback to this state from the observed behavior. We discuss methods for evaluating and maximizing the likelihood of an observed series of behaviors, and thereby estimating parameters, and for inferring the most likely sequence of underlying states. We indicate several extensions of the models, including the incorporation of random effects. We apply these methods in an analysis of the feeding behavior of the caterpillar Helicoverpa armigera, and thereby demonstrate the potential of this family of models as a tool in the investigation of behavior.
    Biometrics 12/2007; 64(3):807-15. · 1.41 Impact Factor
  • Source
    N Neykov, W Zucchini, P Neytchev, H Hristov
    [Show abstract] [Hide abstract]
    ABSTRACT: Non-homogenous hidden Markov Models (NHMMs) have found widespread applica-tion in meteorology and hydrology in Australia and New Zealand, North and South America, e.g., in studies of climate variability or climate change, and in statistical downscaling of daily precipitation from observed and numerical climate model simu-lations., to name just a few.) However, the NHMM methodology has not been yet employed for similar purposes in Europe. The NHMM links large-scale atmospheric patterns to daily precipitation data at a network of rain gauge stations, via several hidden (unobserved) states called the "weather states". The evolution of these states is modelled as a first-order Markov process with state-to-state transition probabilities conditioned on some indices of the atmospheric variables. Due to these weather states the spatial precipitation dependence can be partially or com-pletely captured (see Hughes et al., 1999). In the present study various NHMMs are used to relate daily precipitation at 30 rain gauge stations covering broadly the territory of Bulgaria to synoptic atmospheric data. At each site a 40-year record (1960-2000) of daily October through March precipitation amounts is modelled. The atmospheric data consists of daily sea-level pressure, geopotential height at 500 hPa, air temperature at 850 hPa and relative humidity at 700 hPa on a 2.5 2.5rid based on NCEP-NCAR reanalysis dataset covering the Europe-Atlantic sector 30˚W–60˚E, 20˚N–70˚N for the same period. The first 30 years data are used for model fitting purposes while the re-maining 10 years are used for model evaluation. Detailed model validation is carried out on various aspects. The identified weather states are found to be physically inter-pretable in terms of regional climatology. Acknowledgement: The research of P. Neytchev and N. Neykov is supported by the grant 436 BUL 113/136/0-1 of the Deutsche Forschungsgemeinshchaft within the framework of cooperation between the Deutsche Forschungsgemeinshchaft and the Bulgarian Academy of Sciences.
    Geophysical Research Abstracts. 01/2007; 9(00939).
  • Source
    Hajo Holzmann, Axel Munk, Walter Zucchini
    [Show abstract] [Hide abstract]
    ABSTRACT: We study the issue of identifiability of mixture models in the context of capture-recapture abundance estimation for closed populations. Such models are used to take account of individual heterogeneity in capture probabilities, but their validity was recently questioned by Link (2003, Biometrics 59, 1123-1130) on the basis of their nonidentifiability. We give a general criterion for identifiability of the mixing distribution, and apply it to establish identifiability within families of mixing distributions that are commonly used in this context, including finite and beta mixtures. Our analysis covers binomial and geometrically distributed outcomes. In an example we highlight the difference between the identifiability issue considered here and that in classical binomial mixture models.
    Biometrics 10/2006; 62(3):934-6; discussion 936-9. · 1.41 Impact Factor
  • Walter Zucchini, Iain L. MacDonald
    09/2006; , ISBN: 9780470057339
  • Source
    Hajo Holzmann, Axel Munk, Max Suster, Walter Zucchini
    [Show abstract] [Hide abstract]
    ABSTRACT: We introduce a new class of circular time series based on hidden Markov models. These are compared with existing models, their properties are outlined and issues relating to parameter estimation are discussed. The new models conveniently describe multi-modal circular time series as dependent mixtures of circular distributions. Two examples from biology and meteorology are used to illustrate the theory. Finally, we introduce a hidden Markov model for bivariate linear-circular time series and use it to describe larval movement of the fly Drosophila.
    Environmental and Ecological Statistics 08/2006; 13(3):325-347. · 0.87 Impact Factor
  • Iain L. MacDonald, Walter Zucchini
    08/2006; , ISBN: 9780471667193
  • Source
    Andreas Berzel, Gillian Z. Heller, Walter Zucchini
    [Show abstract] [Hide abstract]
    ABSTRACT: The frequency of doctor consultations has direct consequences for health care budgets, yet little statistical analysis of the determinants of doctor visits has been reported. We consider the distribution of the number of visits to the doctor and, in particular, we model its dependence on a number of demographic factors. Examination of the Australian 1995 National Health Survey data reveals that generalized linear Poisson or negative binomial models are inadequate for modelling the mean as a function of covariates, because of excessive zero counts, and a mean-variance relationship that varies enormously over covariate values. A negative binomial model is used, with parameter values estimated in subgroups according to the discrete combinations of the covariate values. Smoothing splines are then used to smooth and interpolate the parameter values. In effect the mean and the shape parameters are each modelled as (different) functions of gender, age and geographical factors. The estimated regressions for the mean have simple and intuitive interpretations. However, the dependence of the (negative binomial) shape parameter on the covariates is more difficult to interpret and is subject to influence by extreme observations. We illustrate the use of the model by estimating the distribution of the number of doctor consultations in the Statistical Local Area of Ryde, based on population numbers from the 1996 census.
    Australian &amp New Zealand Journal of Statistics 05/2006; 48(2):213 - 224. · 0.53 Impact Factor

Publication Stats

276 Citations
28.81 Total Impact Points

Institutions

  • 1997–2014
    • Georg-August-Universität Göttingen
      • Institute for Mathematical Stochastics
      Göttingen, Lower Saxony, Germany
  • 2006
    • University of Cape Town
      • Department of Actuarial Science
      Cape Town, Province of the Western Cape, South Africa