1995
DOI: 10.1177/0013164495055006001
|View full text |Cite
|
Sign up to set email alerts
|

Use of Empirical Estimates of Shrinkage in Multiple Regression: A Caution

Abstract: Empirical techniques to estimate the shrinkage of the sample R2 have been advocated as alternatives to analytical formulae. Although such techniques may be appropriate for estimating the coefficient of cross-validation, they do not provide accurate estimates of the population multiple correlation. The accuracy of four empirical techniques (simple cross-validation, multi-cross-validation, jackknife, and bootstrap) were investigated in a Monte Carlo study. Random samples of size 20 to 200 were drawn from a pseud… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
34
0

Year Published

2000
2000
2019
2019

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 20 publications
(35 citation statements)
references
References 13 publications
1
34
0
Order By: Relevance
“…Finally, there are various techniques for estimating cross validated population variance explained, r c 2 , such as k-fold cross validation (for a review, see Kromrey & Hines, 1995) …”
Section: Cross-validated R-squaredmentioning
confidence: 99%
“…Finally, there are various techniques for estimating cross validated population variance explained, r c 2 , such as k-fold cross validation (for a review, see Kromrey & Hines, 1995) …”
Section: Cross-validated R-squaredmentioning
confidence: 99%
“…2). Since the sample size and the number of independent variables in the model influence the result of the pCCA (Kromrey & Hines, 1995), Ezekiel's adjustment of fractions was calculated using the following equation (Peres-Neto et al, 2006):…”
Section: Extensively Alteredmentioning
confidence: 99%
“…The variation partitioning algorithm proposed by was used to balance the possible bias in estimation of the variation explained by niche-based (environmental variables) and dispersal-based (geographical variables) processes, due to a different number of explanatory variables being used in these two groups of predictors (Kromrey & Hines 1995). The significance of the pure effects of environmental and geographical variables was then tested by Monte Carlo tests with 1,999 permutations.…”
Section: Statistical Analysesmentioning
confidence: 99%