Minimum volume ellipsoid, Advanced Statistics

Minimum volume ellipsoid is a term for ellipsoid of the minimum volume which covers some specified proportion of the set of multivariate data. It is commonly used to construct robust estimators of the mean vectors and variance-covariance matrices. Such type of estimators has a high breakdown point but is computationally expensive. For instance, for an n × q data matrix X, if h is the integer part of n(n+1)/2 then volumes of n!/(h-n)! Ellipsoids required to be considered to find one with the minimum volume. So for the n=20 there are 184 756 ellipsoids and for the n=30 more than 155 million ellipsoids


Posted Date: 7/30/2012 3:51:59 AM | Location : United States

Related Discussions:- Minimum volume ellipsoid, Assignment Help, Ask Question on Minimum volume ellipsoid, Get Answer, Expert's Help, Minimum volume ellipsoid Discussions

Write discussion on Minimum volume ellipsoid
Your posts are moderated
Related Questions
Advantages and disadvantages of Integrated Economic Statistics

a researcher is interested in whether students who attend privte high schools have higher average SAT Scores than students in the general population. a random sample of 90 student

Quantile regression is an extension of the classical least squares from estimation of the conditional mean models to the estimation of the variety of models for many conditional q

Randomization tests are the procedures for determining the statistical significance directly from the data with- out recourse to some particular sampling distribution. For instanc

Modern hotels and certain establishments make use of an electronic door lock system. To open a door an electronic card is inserted into a slot. A green light indicates that the doo

The graphic representation of the alternatives in a decision making problem which summarizes all the possibilities foreseen by the decision maker. For instance, suppose we are give

The procedures for extracting the pattern in a series of observations when this is obscured by the noise. Basically any such technique or method separates the original series into

Clustered data : The term applied to both the data in which the sampling units are grouped into the clusters sharing some common feature, for instance families or geographical reg

Atomistic fallacy : A fallacy which arises because of the association between two variables at the individual level might vary from the association between the same two variables m

Kalman filter : A recursive procedure which gives an estimate of the signal when only the 'noisy signal' can be observed. The estimate is efficiently constructed by putting the exp