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

Kromrey, Jeffrey D.; Hines, Constance V.

doi:10.1177/0013164495055006001

Cited by 20 publications

(35 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

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%

Incremental criterion prediction of personality facets over factors: Obtaining unbiased estimates and confidence intervals

Anglim

Grant

2014

Journal of Research in Personality

View full text Add to dashboard Cite

Many researchers have argued that higher order models of personality such as the Five Factor Model are insufficient, and that facet-level analysis is required to better understand criteria such as well-being, job performance, and personality disorders. However, common methods in the extant literature used to estimate the incremental prediction of facets over factors have several shortcomings. This paper delineates these shortcomings by evaluating alternative methods using statistical theory, simulation, and an empirical example. We recommend using differences between Olkin-Pratt adjusted r-squared for factor versus facet regression models to estimate the incremental prediction of facets and present a method for obtaining confidence intervals for such estimates using double adjusted-r-squared bootstrapping. We also provide an R package that implements the proposed methods.

show abstract

“…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%

Incremental criterion prediction of personality facets over factors: Obtaining unbiased estimates and confidence intervals

Anglim

Grant

2014

Journal of Research in Personality

View full text Add to dashboard Cite

show abstract

“…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 role of land use, nutrients, and other stressors in shaping benthic invertebrate assemblages in Slovenian rivers

et al. 2011

View full text Add to dashboard Cite

Studying human impacts on riverine systems is challenging due to the natural diversity and multiple stressors that are co-occurring. The term stressor(s) refers to variable(s) of anthropogenic landscape changes and local abiotic stream conditions that reflect human activities. We compared the effects of different stressors and natural (typology) factors on benthic invertebrate assemblages in Slovenian rivers. A total of 270 river sites located in four European ecoregions were sampled, covering the gradient of human perturbation ranging from (near)natural to highly degraded/polluted. Stressors commonly affecting running waters were grouped into managementrelevant stressor-groups: land use (percentage of urban, intensive and non-intensive agricultural, and natural land use in the catchment and sub-catchment), eutrophication (in-stream concentrations of phosphorus and nitrogen compounds) and other stressors (e.g., organic pollution, hydromorphological alteration). We compared the effects of the stressor-groups on invertebrate assemblages using partial Canonical Correspondence Analysis and tested the explanatory power of each stressor-group using typology variables as covariates. Stressors and typology variables showed high independent effects (40-65%) on invertebrate assemblages despite considerable joint effects from land use and typology. Stressors' pure effects accounted for 70%, and interactions among the stressor-groups made up the remaining 30% of the total explained variability. Altogether, this study gives a good perspective on the ability to separate the effects of anthropogenic stressors from natural causes, as well as to separate effects among groups of stressors. Against this background, we advocate stressor-specific analyses to provide or improve guidelines for selecting appropriate measures to enhance the river status.

show abstract

“…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%

Comparison of plant and snail diversity patterns in the White Carpathian Mts (Czech Republic) across forest and grassland habitats

Dvořáková

Horsák

2015

Biologia

View full text Add to dashboard Cite

The Bílé Karpaty Mountains belong to the best-explored areas of the Czech Republic for both land snails and vascular plants, providing a unique opportunity for cross-taxa comparison of their diversity patterns. A total of 1,181 plant and 101 snail species, recorded in 45 grid cells of 2.8 × 3.1 km in size, were used for the analysis. We aimed to investigate the responses of forest and grassland assemblages separately, and to determine relative roles of environmental and spatial predictors on the compositional variation. We observed no significant correlation between the number of plant and snail species across the cells. While land snails showed a negative response to human-made habitats, a positive response was found for plants due to a high proportion of species confined to disturbed and ruderal habitats. Such differences can be attributed to different species pools of forest and grassland species and a high sensitivity of most land snails to habitat degradation and deforestation. In contrast, a positive correlation was found between grassland species as both groups positively responded to the same variables related to habitat diversity. Using the variance partitioning approach, we found a notably higher importance of spatial variables for grassland and forest plant assemblages, in contrast to snail assemblages, which were driven mainly by environmental conditions. These differences emerged especially for grasslands, most likely due to a long-term survival of several relict steppe plants and a notably higher species pool of grassland plants than that of snails.

show abstract

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

Cited by 20 publications

References 13 publications

Incremental criterion prediction of personality facets over factors: Obtaining unbiased estimates and confidence intervals

Incremental criterion prediction of personality facets over factors: Obtaining unbiased estimates and confidence intervals

The role of land use, nutrients, and other stressors in shaping benthic invertebrate assemblages in Slovenian rivers

Comparison of plant and snail diversity patterns in the White Carpathian Mts (Czech Republic) across forest and grassland habitats

Contact Info

Product

Resources

About