feed icon rss

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
Filter
  • Klein, Nadja
  • 1
    UID:
    b3kat_BV049629394
    Format: 1 Online-Ressource (xx, 285 Seiten)
    ISBN: 9781009410076
    Series Statement: Cambridge series in statistical and probabilistic mathematics 56
    Content: An emerging field in statistics, distributional regression facilitates the modelling of the complete conditional distribution, rather than just the mean. This book introduces generalized additive models for location, scale and shape (GAMLSS) - one of the most important classes of distributional regression. Taking a broad perspective, the authors consider penalized likelihood inference, Bayesian inference, and boosting as potential ways of estimating models and illustrate their usage in complex applications. Written by the international team who developed GAMLSS, the text's focus on practical questions and problems sets it apart. Case studies demonstrate how researchers in statistics and other data-rich disciplines can use the model in their work, exploring examples ranging from fetal ultrasounds to social media performance metrics. The R code and data sets for the case studies are available on the book's companion website, allowing for replication and further study.
    Additional Edition: Erscheint auch als Druck-Ausgabe, Hardcover ISBN 978-1-009-41006-9
    Language: English
    URL: Volltext  (URL des Erstveröffentlichers)
    Author information: Kneib, Thomas 1976-
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 2
    UID:
    kobvindex_IGB000024452
    In: Proceedings of the Royal Society of London : Ser. B, Biological Sciences. - 287(2020)1927, art. 20192897
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 3
    UID:
    edochu_18452_27464
    Format: 1 Online-Ressource (18 Seiten)
    ISSN: 0962-2802 , 0962-2802
    Content: We present a new procedure for enhanced variable selection for component-wise gradient boosting. Statistical boosting is a computational approach that emerged from machine learning, which allows to fit regression models in the presence of high-dimensional data. Furthermore, the algorithm can lead to data-driven variable selection. In practice, however, the final models typically tend to include too many variables in some situations. This occurs particularly for low-dimensional data ( p 〈 n), where we observe a slow overfitting behavior of boosting. As a result, more variables get included into the final model without altering the prediction accuracy. Many of these false positives are incorporated with a small coefficient and therefore have a small impact, but lead to a larger model. We try to overcome this issue by giving the algorithm the chance to deselect base-learners with minor importance. We analyze the impact of the new approach on variable selection and prediction performance in comparison to alternative methods including boosting with earlier stopping as well as twin boosting. We illustrate our approach with data of an ongoing cohort study for chronic kidney disease patients, where the most influential predictors for the health-related quality of life measure are selected in a distributional regression approach based on beta regression.
    Content: Peer Reviewed
    Note: This publication is with permission of the rights owner (Sage) freely accessible.
    In: London [u.a.] : Sage Publ., 31,2, Seiten 207-224, 0962-2802
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 4
    UID:
    edochu_18452_27040
    Format: 1 Online-Ressource (12 Seiten)
    Content: Background Statistical model building requires selection of variables for a model depending on the model’s aim. In descriptive and explanatory models, a common recommendation often met in the literature is to include all variables in the model which are assumed or known to be associated with the outcome independent of their identification with data driven selection procedures. An open question is, how reliable this assumed “background knowledge” truly is. In fact, “known” predictors might be findings from preceding studies which may also have employed inappropriate model building strategies. Methods We conducted a simulation study assessing the influence of treating variables as “known predictors” in model building when in fact this knowledge resulting from preceding studies might be insufficient. Within randomly generated preceding study data sets, model building with variable selection was conducted. A variable was subsequently considered as a “known” predictor if a predefined number of preceding studies identified it as relevant. Results Even if several preceding studies identified a variable as a “true” predictor, this classification is often false positive. Moreover, variables not identified might still be truly predictive. This especially holds true if the preceding studies employed inappropriate selection methods such as univariable selection. Conclusions The source of “background knowledge” should be evaluated with care. Knowledge generated on preceding studies can cause misspecification.
    Content: Peer Reviewed
    In: Heidelberg : Springer, 21,1
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 5
    UID:
    edochu_18452_24768
    Format: 1 Online-Ressource (49 Seiten)
    Content: In dieser Arbeit wird eine Methode für die Schätzung von elastischen Formmittelwerten von offenen und planaren Kurven vorgestellt. Dazu wurden die Konzepte des full Procrustes Mittelwerts für Landmark Konfigurationen mit dem square-root-velocity Framework für die elastische Analyse von Funktionen kombiniert, um elastische full Procrustes Mittelwerte zu berechnen. Die Schätzung dieses Mittelwerts hängt eng mit der komplexen Kovarianzoberfläche der beobachteten Kurven zusammen, welche durch Berücksichtigung ihrer Symmetrieeigenschaften effizient geschätzt werden kann (siehe Cederbaum, Scheipl und Greven 2018). In Kombination mit den Methoden für elastische Mittelwertberechnung in Steyer, Stöcker und Greven 2021 ist dieser Ansatz besonders passend Kurven mit nur wenigen, unregelmäßig verteilten Beobachtungspunkten. Die vorgestellte Methode und ihre Implementierung wurde auf zwei Testdatensätzen überprüft. Über den in der Schätzung verwendeten Strafparameter (penalty) kann die Glätte des Mittelwerts kontrolliert werden, wobei eine penalty erster oder zweiter Ordnung auch in Bereichen mit wenig Beobachtungen zu stabilen Schätzungen führt. Die Beobachtungen können über ihre elastischen full Procrustes Fits gemeinsam mit dem Mittelwert visualisiert werde, wobei der geschätzte Mittelwert aber nicht als ein naiver funktionaler Mittelwert der Fits interpretiert werden kann. Außerdem ist der elastische full Procrustes Mittelwert robust gegenüber Ausreißern, da diese tendenziell so skaliert werden, das ihr Einfluss auf die Schätzung eingeschränkt wird. Die Methode wurde benutzt um die Variabilität von Zungenkonturen bei der Artikulation von Konsonanten zu analysieren. Hierbei scheint der Vokalkontext den größten Effekt auf die Zungenkontur zu haben, während der Einfluss des ausgesprochen Konsonanten zwischen den Sprechern unterscheidet.
    Content: This thesis proposes a method for elastic shape mean estimation of open planar curves. It combines the concept of the full Procrustes mean with the square-root-velocity framework to estimate elastic full Procrustes means. This mean can be shown to be related to an eigenfunction problem over the complex covariance surface of the observed curves, which can be estimated efficiently by accounting for its symmetry (Cederbaum, Scheipl and Greven 2018). These methods, in combination with the methods for elastic mean estimation in Steyer, Stöcker and Greven 2021, make the proposed estimation procedure especially suitable to sparse and irregular observations. The method and its implementation were validated over toy datasets, where it was shown that the estimated shape mean is elastic and invariant with respect to shape preserving transformations of the input curves. The penalty used in the estimation provides stable estimates in regions where observations are sparse and consistently good results were achieved for first and second order penalties. The estimated mean may be plotted together with the elastic full Procrustes fits, however, care has to be taken when interpreting these plots, as the estimated mean curve is not a simple functional mean of the aligned curves. Furthermore, the elastic full Procrustes mean is quite robust to outliers, as they tend to get shrunk towards zero, restricting their influence on the mean estimation. Finally, the method was used to analyse variability in an empirical dataset of tongue contours during consonant articulation. Here it was found that the vowel context has the largest influence on the tongues shape, while the effect of the spoken consonant seems to vary between speakers.
    Note: Masterarbeit Humboldt-Universität zu Berlin 2021
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 6
    Online Resource
    Online Resource
    Berlin : Humboldt-Universität zu Berlin
    UID:
    edochu_18452_26113
    Format: 1 Online-Ressource (20 Seiten)
    Content: Spatial models are used in a variety of research areas, such as environmental sciences, epidemiology, or physics. A common phenomenon in such spatial regression models is spatial confounding. This phenomenon is observed when spatially indexed covariates modeling the mean of the response are correlated with a spatial random effect included in the model, for example, as a proxy of unobserved spatial confounders. As a result, estimates for regression coefficients of the covariates can be severely biased and interpretation of these is no longer valid. Recent literature has shown that typical solutions for reducing spatial confounding can lead to misleading and counterintuitive results. In this article, we develop a computationally efficient spatial model that explicitly correlates a Gaussian random field for the covariate of interest with the Gaussian random field in the main model equation and integrates novel prior structures to reduce spatial confounding. Starting from the univariate case, we extend our prior structure also to the case of multiple spatially confounded covariates. In simulation studies, we show that our novel model flexibly detects and reduces spatial confounding in spatial datasets, and it performs better than typically used methods such as restricted spatial regression. These results are promising for any applied researcher who wishes to interpret covariate effects in spatial regression models. As a real data illustration, we study the effect of elevation and temperature on the mean of monthly precipitation in Germany.
    Content: Peer Reviewed
    In: Chichester, West Sussex : Wiley, 33,5
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 7
    UID:
    edochu_18452_28420
    Format: 1 Online-Ressource (13 Seiten)
    Content: Capturing complex dependence structures between outcome variables (e.g., study endpoints) is of high relevance in contemporary biomedical data problems and medical research. Distributional copula regression provides a flexible tool to model the joint distribution of multiple outcome variables by disentangling the marginal response distributions and their dependence structure. In a regression setup, each parameter of the copula model, that is, the marginal distribution parameters and the copula dependence parameters, can be related to covariates via structured additive predictors. We propose a framework to fit distributional copula regression via model‐based boosting, which is a modern estimation technique that incorporates useful features like an intrinsic variable selection mechanism, parameter shrinkage and the capability to fit regression models in high‐dimensional data setting, that is, situations with more covariates than observations. Thus, model‐based boosting does not only complement existing Bayesian and maximum‐likelihood based estimation frameworks for this model class but rather enables unique intrinsic mechanisms that can be helpful in many applied problems. The performance of our boosting algorithm for copula regression models with continuous margins is evaluated in simulation studies that cover low‐ and high‐dimensional data settings and situations with and without dependence between the responses. Moreover, distributional copula boosting is used to jointly analyze and predict the length and the weight of newborns conditional on sonographic measurements of the fetus before delivery together with other clinical variables.
    Content: Peer Reviewed
    In: Malden, Mass. : Wiley-Blackwell, 79,3, Seiten 2298-2310
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 8
    Online Resource
    Online Resource
    Berlin : Humboldt-Universität zu Berlin
    UID:
    edochu_18452_29179
    Format: 1 Online-Ressource (18 Seiten)
    Content: Gaussian mixture models are a popular tool for model-based clustering, and mixtures of factor analyzers are Gaussian mixture models having parsimonious factor covariance structure for mixture components. There are several recent extensions of mixture of factor analyzers to deep mixtures, where the Gaussian model for the latent factors is replaced by a mixture of factor analyzers. This construction can be iterated to obtain a model with many layers. These deep models are challenging to fit, and we consider Bayesian inference using sparsity priors to further regularize the estimation. A scalable natural gradient variational inference algorithm is developed for fitting the model, and we suggest computationally efficient approaches to the architecture choice using overfitted mixtures where unnecessary components drop out in the estimation. In a number of simulated and two real examples, we demonstrate the versatility of our approach for high-dimensional problems, and demonstrate that the use of sparsity inducing priors can be helpful for obtaining improved clustering results.
    Content: Peer Reviewed
    In: Dordrecht [u.a.] : Springer Science + Business Media B.V, 32,5
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
  • 9
    Online Resource
    Online Resource
    Berlin : Humboldt-Universität zu Berlin
    UID:
    edochu_18452_29374
    Format: 1 Online-Ressource (29 Seiten)
    Content: We propose three novel consistent specification tests for quantile regression models which generalize former tests in three ways. First, we allow the covariate effects to be quantile-dependent and nonlinear. Second, we allow parameterizing the conditional quantile functions by appropriate basis functions, rather than parametrically. We are thereby able to test for general functional forms, while retaining linear effects as special cases. In both cases, the induced class of conditional distribution functions is tested with a Cramér–von Mises type test statistic for which we derive the theoretical limit distribution and propose a bootstrap method. Third, a modified test statistic is derived to increase the power of the tests. We highlight the merits of our tests in a detailed MC study and two real data examples. Our first application to conditional income distributions in Germany indicates that there are not only still significant differences between East and West but also across the quantiles of the conditional income distributions, when conditioning on age and year. The second application to data from the Australian national electricity market reveals the importance of using interaction effects for modeling the highly skewed and heavy-tailed distributions of energy prices conditional on day, time of day and demand.
    Content: Peer Reviewed
    In: Oxford : Wiley-Blackwell, 51,1, Seiten 355-383
    Language: English
    URL: Volltext  (kostenfrei)
    Library Location Call Number Volume/Issue/Year Availability
    BibTip Others were also interested in ...
Close ⊗
This website uses cookies and the analysis tool Matomo. Further information can be found on the KOBV privacy pages