CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Traceable Regressions

Nanny Wermuth (Institutionen för matematiska vetenskaper, matematisk statistik)
International Statistical Review (0306-7734). Vol. 80 (2012), 3, p. 415-438.
[Artikel, refereegranskad vetenskaplig]

In this paper, we define and study the concept of traceable regressions and apply it to some examples. Traceable regressions are sequences of conditional distributions in joint or single responses for which a corresponding graph captures an independence structure and represents, in addition, conditional dependences that permit the tracing of pathways of dependence. We give the properties needed for transforming these graphs and graphical criteria to decide whether a path in the graph induces a dependence. The much stronger constraints on distributions that are faithful to a graph are compared to those needed for traceable regressions.

Nyckelord: Chain graphs, edge-matrix calculus, faithfulness of graphs, graphical Markov models, independence, roc curve, classifier performance, diagnostic-tests, scoring rules, area, prediction, overlap



Denna post skapades 2012-12-20. Senast ändrad 2016-07-12.
CPL Pubid: 168446

 

Läs direkt!


Länk till annan sajt (kan kräva inloggning)


Institutioner (Chalmers)

Institutionen för matematiska vetenskaper, matematisk statistik (2005-2016)

Ämnesområden

Matematik

Chalmers infrastruktur