Multivariate Ordered Logit Regressions
نویسنده
چکیده
In this paper we combine recent advances in marginal modelling for contingency tables with the notion of copula to formulate a class of models for describing how the joint distribution of a set of ordinal response variables depends on exogenous regressors. We derive the main properties of a marginal parameterization, the global interaction copula, whose nature is essentially non parametric, and study its relation with an underlying seemingly unrelated system of latent variables regression. An efficient algorithm for maximum likelihood estimation is described; we also derive the asymptotic distribution of the likelihood ratio under suitable equality and inequality constraints. An application to O-levels grades for a cohort of UK students clarifies the usefulness of these models. JEL codes: C35, C12, C13.
منابع مشابه
Working Paper Series Categorical Data Categorical Data
Categorical outcome (or discrete outcome or qualitative response) regression models are models for a discrete dependent variable recording in which of two or more categories an outcome of interest lies. For binary data (two categories) probit and logit models or semiparametric methods are used. For multinomial data (more than two categories) that are unordered, common models are multinomial and...
متن کاملHow Important Is Methodology for the Estimates of the Determinants of Happiness?*
Psychologists and sociologists usually interpret happiness scores as cardinal and comparable across respondents, and thus run OLS regressions on happiness and changes in happiness. Economists usually assume only ordinality and have mainly used ordered latent response models, thereby not taking satisfactory account of fixed individual traits. We address this problem by developing a conditional e...
متن کاملUnivariate or Multivariate Analysis for Better Prediction Accuracy ? A Case Study of Heterogeneity in Vehicle Ownership
In this research, we contrast different modeling frameworks that offer alternative ways of capturing observed/unobserved heterogeneity. The model systems compared are: ordered logit, residential location cluster based ordered logit model (exogenous segmentation), mixed ordered logit, latent segmentation based ordered logit model, and a joint copula based self-selection clustering model. While t...
متن کاملHedging And The Use Of Derivatives: Evidence From UK Non-Financial Firms
This paper attempts to differentiate among the theories of corporate hedging by using treasury disclosures in the 1995 annual reports of the largest 400 UK companies and data collected via a survey to corporate treasurers. Univariate and multivariate logit tests indicate that firms with tax loss carryforwards, higher gearing, lower liquidity, foreign exchange exposure and larger firms are more ...
متن کاملThe Use and Impact of Business Advice by Smes in Britain: an Empirical Assessment Using Logit and Ordered Logit
This paper assesses the effect of differences in types of client on the use and impact of business advice by SMEs in Britain using new survey evidence from the Cambridge ESRC Centre for Business Research Survey of 1997. The survey, covering over 2500 respondents, is the largest and most definitive assessment available in Britain. Moreover, the survey allows an assessment of the full range of th...
متن کامل