A proportional hazard cure model for ordinal responses by self-modeling regression

ABSTRACT

In a medical study, patients have various stages of illness. After treatment the patient will be cured or the stage of illness will change. Since there are suitable evidences of a susceptible population by several levels, the authors combine a Self-Modeling ordinal model for the probability of occurrence of an event with a Cox regression for the time of occurrence of an event. We proposed the use of self-modeling ordinal longitudinal where the conditional cumulative probabilities for a category of an outcome have a relation with shape-invariant model. A simulation study is carried out for justification of the methodology. A schizophrenia illness data are analyzed based on our model to see whether the treatment affects the illness.

KEYWORDS:

• Proportional hazard model
• self-modeling
• ordinal response
• Monte Carlo method

Acknowledgments

We would like to express our sincere thanks to the reviewers for their valuable comments that has improved the revised manuscript to a great extent. Mastani Shirazi heartily thanks the generous support of CAPES-Brazil.

Disclosure statement

No potential conflict of interest was reported by the authors.