Point scoring, Advanced Statistics

Point scoring is an easy distribution free method which can be used for the prediction of a response which is a binary variable from the observations on several explanatory variables which are also binary in nature. The easiest version of the process, often known as the Burgess technique, operates by first taking the explanatory variables one at a time and then determining which level of each variable is related with the higher proportion of 'success' category of the binary response. The prediction score for any of the individual is then just the number of explanatory variables at the high level (generally only variables which are ' significant' are included in the score). The score thus varies from 0, when all explanatory variables are at low level, to its maximum value when all important variables are at the high level. The goal of the technique is to split the population into risk groups.

Posted Date: 7/31/2012 1:45:21 AM | Location : United States







Related Discussions:- Point scoring, Assignment Help, Ask Question on Point scoring, Get Answer, Expert's Help, Point scoring Discussions

Write discussion on Point scoring
Your posts are moderated
Related Questions
Classification matrix: A term many times used in discriminant analysis for the matrix summarizing the results and outputs obtained from the derived classi?cation rule, and obtaine

Longitudinal data : The data arising when each of the number of subjects or patients give rise to the vector of measurements representing same variable observed at the number of di

Kurtosis: The extent to which the peak of the unimodal probability distribution or the frequency distribution departs from its shape of the normal distribution, by either being mo

Mortality odds ratio  is the ratio equivalent to the odds ratio used in case-control studies where the equivalent of the cases are deaths from the cause of interest and the equival

Non linear mapping (NLM ) is a technique for obtaining a low-dimensional representation of the set of multivariate data, which operates by minimizing a function of the differences

What is the EM?

Individual differences scaling is a form of multidimensional scaling applicable to the data comprising of a number of proximity matrices from the different sources that is differe

The method or technique for producing the sequence of parameter estimates that, under the mild regularity conditions, converges to maximum likelihood estimator. Of particular signi

Collective risk models : The models applied to insurance portfolios which do not create direct reference to the risk characteristics of individual members of the portfolio when des

Laplace distribution : The probability distribution, f(x), given by the following formula   Can be derived as the distribution of the difference of two independent random var