Graphical Methods for Assessing Logistic Regression Models

Landwehr, James M.; Pregibon, Daryl; Shoemaker, Anne C.

doi:10.1080/01621459.1984.10477062

Cited by 249 publications

(77 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One way to cope with the issue is to cluster individuals based on the explanatory variables and to extract "pure error term" from constructing "pseudo-replications" on each cluster. This idea is an extention of suggestions by Tsiatis (1980), Fienberg & Gong (1984) and Landwehr et al (1984) developed in the case of binary response, namely the logistic model, to multi responses. Another way to consider the problem is to evaluate the dependency of the ordered categorical response on the explanatory variables, such as discrimination problem among "ordered populations" .…”

Section: Interpretation Of the Results Of Analysismentioning

confidence: 80%

“…If we focus on the analysis of the logistic model, fairly abundant tools for the diagnoses have been developed (Pregibon, 1981 ;Schoenfeld, 1982;Landwehr et al, 1984). We briefly mention these tools for diagnosis about the model (2) based on the APT.…”

Section: Discussionmentioning

confidence: 99%

“…Further if we wish to evaluate the proportion of variations explained (PVE) by the explanatory variables, we may apply the conditional likelihood ratio test proposed by Fienberg & Gong (1984), or score test by Tsiatis (1980). To check the goodness of fit, graphical method are most useful, and some methods such as local mean deviance plotting by Landwehr et al (1984) can be applied. The plotting may be effective to detect interaction between the explana tory variables.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Power-Transformation for Ordered Categorical Data

1986

View full text Add to dashboard Cite

We consider in this paper a model of asymmetric power-transformation of response probability explained by linear function of some covariates. This model includes logistic and complementary log-log transformation models as its specific case, so we can use the model to evaluate the appropriateness or the goodness of fit of these models. Then the performances of the asymmetric power-transformation model are evaluated and examined, based on data used in published literatures.And we discuss various issues of diagnoses which may occur in the process of applying the model. Further we consider alternative transformations, and then in comparison with them, we point out the advantages of the asymmetric power-transformation.

show abstract

Section: Interpretation Of the Results Of Analysismentioning

confidence: 80%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Power-Transformation for Ordered Categorical Data

1986

View full text Add to dashboard Cite

show abstract

“…However, the deviance given by (9) is not even approximately a χ 2 distribution for the case in which ungrouped binary responses are available (e.g., see Collett, 2003;Landwehr et al, 1984;Tsujitani and Koshimizu, 2000;Tsujitani and Sakon, 2009). The number of degrees of freedom required for the test for significance using the assumed χ 2 distribution for the deviance is a contentious issue.…”

Section: Generalized Additive Modelsmentioning

confidence: 99%

Survival Analysis for Personal Loan Data Using Generalized Additive Models

Tsujitani¹,

Baesens

2012

Behaviormetrika

View full text Add to dashboard Cite

This article presents flexible methods for modeling censored survival data using penalized smoothing splines when the covariate values change for the duration of the study. The Cox proportional hazards model has been widely used for the analysis of censored survival data. However, a number of theoretical problems with respect to the baseline survival function and the baseline cumulative hazard function remain unsolved. The basic concept considered in the present article is to use generalized additive models (GAM) with B-splines to estimate the survival function without the baseline hazard assumption. The proposed methods are discussed according to the way in which they deal with censored observations, competing risk, and time-dependent covariates. We evaluate the performance of the proposed method for predicting loan default with early payment as competing risk using data from a U.K. financial institution.

show abstract

“…Unlike the diagnostics for linear regressions in Reiter (2003), which are based on unit-specific residuals, the categorical data regression diagnostics are based on grouped residuals (Landwehr et al, 1984;Gelman et al, 2000). The article is organized as follows.…”

Section: Introductionmentioning

confidence: 99%

Categorical data regression diagnostics for remote access servers

Reiter

Kohnen

2005

Journal of Statistical Computation and Simulation

View full text Add to dashboard Cite

Owing to the growing concerns over data confidentiality, many national statistical agencies are considering remote access servers to disseminate data to the public. With remote servers, users submit requests for output from statistical models fit using the collected data, but they are not allowed access to the data. Remote servers also should enable users to check the fit of their models; however, standard diagnostics like residuals or influence statistics can disclose individual data values. In this article, we present diagnostics for categorical data regressions that can be safely and usefully employed in remote servers. We illustrate the diagnostics with simulation studies.

show abstract

Graphical Methods for Assessing Logistic Regression Models

Cited by 249 publications

References 7 publications

Power-Transformation for Ordered Categorical Data

Power-Transformation for Ordered Categorical Data

Survival Analysis for Personal Loan Data Using Generalized Additive Models

Categorical data regression diagnostics for remote access servers

Contact Info

Product

Resources

About