2018
DOI: 10.18637/jss.v087.i09
|View full text |Cite
|
Sign up to set email alerts
|

Visualizing Fit and Lack of Fit in Complex Regression Models with Predictor Effect Plots and Partial Residuals

Abstract: Predictor effect displays, introduced in this article, visualize the response surface of complex regression models by averaging and conditioning, producing a sequence of 2D line graphs, one graph or set of graphs for each predictor in the regression problem. Partial residual plots visualize lack of fit, traditionally in relatively simple additive regression models. We combine partial residuals with effect displays to visualize both fit and lack of fit simultaneously in complex regression models, plotting resid… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

1
285
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
10

Relationship

0
10

Authors

Journals

citations
Cited by 432 publications
(286 citation statements)
references
References 21 publications
1
285
0
Order By: Relevance
“…Corresponding negative binomial or Poisson models did not fit the data adequately. Results were plotted using the effects package (Fox and Weisberg 2018).…”
Section: Statisticsmentioning
confidence: 99%
“…Corresponding negative binomial or Poisson models did not fit the data adequately. Results were plotted using the effects package (Fox and Weisberg 2018).…”
Section: Statisticsmentioning
confidence: 99%
“…, C × S, β‐std = −1.32, P = 0.07; Fig. , 95% CI show GLM fit, following Fox & Weisberg, ). In contrast, sap‐feeders had a negative effect on shoot growth only when ants were present (Fig.…”
Section: Resultsmentioning
confidence: 88%
“…The package lmerTest [50] was used to calculate standard errors, effect sizes, and significance values. The effects of the models were visualized in plots for a better interpretation of each model by applying the effects package [51].…”
Section: Resultsmentioning
confidence: 99%