What question did this study set out to answer?

To formulate and evaluate partial proportional odds models that relax the proportional odds assumption for select explanatory variables in ordinal regression.

January 1, 1990Journal of the Royal Statistical Society Series C (Applied Statistics)

Partial Proportional Odds Models for Ordinal Response Variables

View Full Paper

Ask AI

Bookmark

To formulate and evaluate partial proportional odds models that relax the proportional odds assumption for select explanatory variables in ordinal regression.
Extended cumulative logit models to permit covariate-specific relaxation of equal-slope constraints across outcome thresholds.
Applied maximum likelihood estimation procedures and score tests to assess proportional odds violations and estimate variable effects.
Demonstrated that partial proportional odds models retain ordinal information without the parameter proliferation seen in nominal multinomial logit models.
Showed improved model specification and parameter interpretability when only a subset of predictors violated the proportional odds assumption.

Implication

Methodological analysis reveals partial proportional odds modeling for ordinal outcomes, indicating improved flexibility when proportional odds assumptions fail.

Abstract

Bercedis Peterson, Frank E. Harrell, Jr, Partial Proportional Odds Models for Ordinal Response Variables, Journal of the Royal Statistical Society. Series C (Applied Statistics), Vol. 39, No. 2 (1990), pp. 205-217

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Peterson et al. (1990) studied this question.

synapsesocial.com/papers/69d99a790d540cafc583643e https://doi.org/10.2307/2347760

View Full Paper

Ask AI

Bookmark

Implication

Methodological analysis reveals partial proportional odds modeling for ordinal outcomes, indicating improved flexibility when proportional odds assumptions fail.

Partial Proportional Odds Models for Ordinal Response Variables

Key Points

Implication

Abstract

Discussion

Cite This Study

Partial Proportional Odds Models for Ordinal Response Variables

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading