podcast.de
Anmeldung
Anmeldung Registrierung
Auto Hell Dunkel
Erweiterte Suche
  1. Startseite
  2. Podcasts
  3. Mathematik, Informatik und Statistik - Open Access LMU - Teil 03/03 Podcast
Mathematik, Informatik und Statistik - Open Access LMU - Teil 03/03

Mathematik, Informatik und Statistik - Open Access LMU - Teil 03/03

Ludwig-Maximilians-Universität München 50 Episoden Neueste vor 10 Jahren
Bildung
Zu Sammlung hinzufügen

Du hast noch keine Sammlungen.
Teilen Kommentare Alle Episoden anzeigen

Episoden

A General Framework for the Selection of Effect Type in Ordinal Regression 2/2

1 minute
In regression models for ordinal response, each covariate can be equipped with either a simple, global effect or a more flexible and complex effect which is specific to the response categories. Instead of a priori assuming one of these effect types, as is done in the majority of the literature, we argue in this paper that effect type selection shall be data-based. For this purpose, we propose a novel and general penalty framework that allows for an automatic, data-driven selection between global and category-specific effects in all types of ordinal regression models. Optimality conditions and an estimation algorithm for the resulting penalized estimator are given. We show that our approach is asymptotically consistent in both effect type and variable selection and possesses the oracle property. A detailed application further illustrates the workings of our method and demonstrates the advantages of effect type selection on real data.

A General Framework for the Selection of Effect Type in Ordinal Regression 1/2

1 minute
In regression models for ordinal response, each covariate can be equipped with either a simple, global effect or a more flexible and complex effect which is specific to the response categories. Instead of a priori assuming one of these effect types, as is done in the majority of the literature, we argue in this paper that effect type selection shall be data-based. For this purpose, we propose a novel and general penalty framework that allows for an automatic, data-driven selection between global and category-specific effects in all types of ordinal regression models. Optimality conditions and an estimation algorithm for the resulting penalized estimator are given. We show that our approach is asymptotically consistent in both effect type and variable selection and possesses the oracle property. A detailed application further illustrates the workings of our method and demonstrates the advantages of effect type selection on real data.

Identifiability in penalized function-on-function regression models

1 minute
Regression models with functional responses and covariates constitute a powerful and increasingly important model class. However, regression with functional data poses well known and challenging problems of non-identifiability. This non-identifiability can manifest itself in arbitrarily large errors for coefficient surface estimates despite accurate predictions of the responses, thus invalidating substantial interpretations of the fitted models. We offer an accessible rephrasing of these identifiability issues in realistic applications of penalized linear function-on-function-regression and delimit the set of circumstances under which they are likely to occur in practice. Specifically, non-identifiability that persists under smoothness assumptions on the coefficient surface can occur if the functional covariate's empirical covariance has a kernel which overlaps that of the roughness penalty of the spline estimator. Extensive simulation studies validate the theoretical insights, explore the extent of the problem and allow us to evaluate their practical consequences under varying assumptions about the data generating processes. A case study illustrates the practical significance of the problem. Based on theoretical considerations and our empirical evaluation, we provide immediately applicable diagnostics for lack of identifiability and give recommendations for avoiding estimation artifacts in practice.

Identifiability in penalized function-on-function regression models

1 minute
Regression models with functional covariates for functional responses constitute a powerful and increasingly important model class. However, regression with functional data poses challenging problems of non-identifiability. We describe these identifiability issues in realistic applications of penalized linear function-on-function-regression and delimit the set of circumstances under which they arise. Specifically, functional covariates whose empirical covariance has lower effective rank than the number of marginal basis function used to represent the coefficient surface can lead to unidentifiability. Extensive simulation studies validate the theoretical insights, explore the extent of the problem and allow us to evaluate its practical consequences under varying assumptions about the data generating processes. Based on theoretical considerations and our empirical evaluation, we provide easily verifiable criteria for lack of identifiability and provide actionable advice for avoiding spurious estimation artifacts. Applicability of our strategy for mitigating non-identifiability is demonstrated in a case study on the Canadian Weather data set.

Evaluation of a new k-means approach for exploratory clustering of items

1 minute
Evaluation of a new k-means approach for exploratory clustering of items
Alle Folgen ansehen (50)
In Podcasts werben
Schalte jetzt Werbung in Podcasts.

Über diesen Podcast

Die Universitätsbibliothek (UB) verfügt über ein umfangreiches Archiv an elektronischen Medien, das von Volltextsammlungen über Zeitungsarchive, Wörterbücher und Enzyklopädien bis hin zu ausführlichen Bibliographien und mehr als 1000 Datenbanken reicht. Auf iTunes U stellt die UB unter anderem eine Auswahl an elektronischen Publikationen der Wissenschaftlerinnen und Wissenschaftler an der LMU bereit. (Dies ist der 3. von 3 Teilen der Sammlung 'Mathematik, Informatik und Statistik - Open Access LMU'.)
Webseite
Open Access LMU
Urheberangabe
Ludwig-Maximilians-Universität München
Erscheinungsweise
Unregelmäßig
Letzter Check
vor 7 Std.
Zuerst veröffentlicht
vor 10 J.
Podcast melden

„Mathematik, Informatik und Statistik - Open Access LMU - Teil 03/03“

Worum geht es? Danach fragen wir noch nach dem Grund.

Abonnenten

Teilen

Mein Archiv

Deine Privatkopie der Folgen, die du nicht verlieren willst.

Podcast-Folgen verschwinden. Feeds werden auf die letzten Episoden gekürzt, Hoster räumen alte Dateien ab, Formate wechseln den Anbieter und lassen ihr Archiv zurück. Mit „Mein Archiv“ sichert podcast.de die Folgen deiner Podcasts für dich — angefangen bei den ältesten, denn die sind zuerst weg.

  • Deine gesicherten Folgen bleiben hörbar, auch wenn das Original offline geht.
  • Auch Folgen, die im heutigen Feed gar nicht mehr stehen — podcast.de kennt sie noch.
  • Herunterladen bleibt möglich, solange die Folge beim Podcaster liegt. Der zählt seine Abrufe wie bisher.
Startet bald

Sei beim Start von Mein Archiv dabei

Mein Archiv ist fast fertig. Trag dich ein, dann bekommst du eine E-Mail, sobald es losgeht – und bist von Anfang an dabei. Wir schreiben dir nur zum Start, keine Werbung, keine Weitergabe deiner Daten.

Du bekommst zuerst eine Bestätigungsmail. Abmelden geht jederzeit. Datenschutz