Projection pursuit, Advanced Statistics

Projection pursuit is a procedure for attaning a low-dimensional (usually two-dimensional) representation of the multivariate data, which will be particularly useful in revealing the interesting structure such as presence of distinct groups of the observations. A low-dimensional representation is found by optimizing some pre-defined numerical criterion designed to provide with 'interesting' patterns.

Posted Date: 7/31/2012 6:27:30 AM | Location : United States







Related Discussions:- Projection pursuit, Assignment Help, Ask Question on Projection pursuit, Get Answer, Expert's Help, Projection pursuit Discussions

Write discussion on Projection pursuit
Your posts are moderated
Related Questions
Johnson''s Job Sequencing for n jobs and 2 machines

Helmert contrast is the contrast often used in analysis of the variance, in which each level of a factor is tested against average of the remaining levels. So, for instance, if th

Given: There are 4 jobs and 4 persons. The cost incurred for each person and each job is as follows: Persons Job 1 Job 2 Job 3 Job 4 A 10 9 21 11 B 15 12 25 17 C 12 10 20 12 D 17

An approach of using the likelihood as the basis of estimation without the requirement to specify a parametric family for data. Empirical likelihood can be viewed as the example of

Data which occur when failure period is recorded which are dependent. Such type of data can arise in number contexts, for instance, in epidemiological cohort studies in which th

Computer-intensive methods : The statistical methods which require almost identical computations on the data repeated number of times. The term computer intensive is, certainly, a

Chi-squared distribution : It is the probability distribution, f (x), of the random variable de?ned as the sum of squares of the number (v) of independent standard normal variables

cholscores Treatment income ($000) Patient ID low Income? 0.6 Old 21.3 2 Yes 0.17 Old 27.2 13 Yes 0.69 New 27.1 16 Yes 1.09 Old 94.8

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

This is extension of the EM algorithm which typically converges more slowly than EM in terms of the iterations but can be much faster in the whole computer time. The general idea o