1984
DOI: 10.2307/2981683
|View full text |Cite
|
Sign up to set email alerts
|

Present Position and Potential Developments: Some Personal Views: Statistical Theory: The Prequential Approach

Abstract: SUMMARY The prequential approach is founded on the premiss that the purpose of statistical inference is to make sequential probability forecasts for future observations, rather than to express information about parameters. Many traditional parametric concepts, such as consistency and efficiency, prove to have natural counterparts in this formulation, which sheds new light on these and suggests fruitful extensions.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

3
366
0
3

Year Published

2000
2000
2016
2016

Publication Types

Select...
6
2
1

Relationship

0
9

Authors

Journals

citations
Cited by 888 publications
(372 citation statements)
references
References 18 publications
3
366
0
3
Order By: Relevance
“…compare MDL and MML from a predictive point of view. The predictive interpretation of MDL has close analogies with Dawid's prequential principle (Dawid 1984; and a version of cross-validation called forward validation (Rissanen 1989, Section 3.3). In some cases, it is not necessary to pick a specific model order; it may be better, in fact, to employ all the models in a weighted fashion (Wasserman, 2000;Hoeting et al, 1999;Chickering & Heckerman, 2000), as Singer & Feder (1999) illustrate for linear least-squares prediction problems.…”
Section: Resultsmentioning
confidence: 98%
“…compare MDL and MML from a predictive point of view. The predictive interpretation of MDL has close analogies with Dawid's prequential principle (Dawid 1984; and a version of cross-validation called forward validation (Rissanen 1989, Section 3.3). In some cases, it is not necessary to pick a specific model order; it may be better, in fact, to employ all the models in a weighted fashion (Wasserman, 2000;Hoeting et al, 1999;Chickering & Heckerman, 2000), as Singer & Feder (1999) illustrate for linear least-squares prediction problems.…”
Section: Resultsmentioning
confidence: 98%
“…Given the probabilities generated by each individual predictor, P(Z t = 1|X it ) for i = 1, 2, ..., I, all of which may be perfectly calibrated in the sense of Dawid (1984) 4 , the linearly combined forecast can be constructed by taking the linear pool, that is,…”
Section: Relationship With Extant Methodsmentioning
confidence: 99%
“…Having witnessed a finite sequence σ of bits, we are to make a probability forecast, based on σ only, of what bit comes next; then this bit is revealed, and the procedure is repeated. Dawid (1984) names it the prequential approach, for sequential prediction in a probabilistic fashion.…”
Section: Setting the Stagementioning
confidence: 99%