In medical, biological and social science, three dimensional contingency tables with ordered categories usually arise and one expects a monotone association yet one may prefer not to assume a particular parametric model. One of its key roles is how to estimate its parameters under monotonic...

type="main" xml:id="sjos12057-abs-0001" <title type="main">ABSTRACT</title>Motivated by an entropy inequality, we propose for the first time a penalized profile likelihood method for simultaneously selecting significant variables and estimating unknown coefficients in multiple linear regression models in this article. The...