Estimation in A General Context
Given a sample consisting of n independent observations x1,..., xn of a p-dimensional random vector X ∈ Rp×1 (a p×1 column-vector), an unbiased estimator of the (p×p) covariance matrix
is the sample covariance matrix
where is the i-th observation of the p-dimensional random vector, and
is the sample mean. This is true regardless of the distribution of the random variable X, provided of course that the theoretical means and covariances exist. The reason for the factor n − 1 rather than n is essentially the same as the reason for the same factor appearing in unbiased estimates of sample variances and sample covariances, which relates to the fact that the mean is not known and is replaced by the sample mean.
In cases where the distribution of the random variable X is known to be within a certain family of distributions, other estimates may be derived on the basis of that assumption. A well-known instance is when the random variable X is normally distributed: in this case the maximum likelihood estimator of the covariance matrix is slightly different from the unbiased estimate, and is given by
A derivation of this result is given below. Clearly, the difference between the unbiased estimator and the maximum likelihood estimator diminishes for large n.
In the general case, the unbiased estimate of the covariance matrix provides an acceptable estimate when the data vectors in the observed data set are all complete: that is they contain no missing elements. One approach to estimating the covariance matrix is to treat the estimation of each variance or pairwise covariance separately, and to use all the observations for which both variables have valid values. Assuming the missing data are missing at random this results in an estimate for the covariance matrix which is unbiased. However, for many applications this estimate may not be acceptable because the estimated covariance matrix is not guaranteed to be positive semi-definite. This could lead to estimated correlations having absolute values which are greater than one, and/or a non-invertible covariance matrix.
Read more about this topic: Estimation Of Covariance Matrices
Famous quotes containing the words estimation, general and/or context:
“No man ever stood lower in my estimation for having a patch in his clothes; yet I am sure that there is greater anxiety, commonly, to have fashionable, or at least clean and unpatched clothes, than to have a sound conscience.”
—Henry David Thoreau (1817–1862)
“I have never looked at foreign countries or gone there but with the purpose of getting to know the general human qualities that are spread all over the earth in very different forms, and then to find these qualities again in my own country and to recognize and to further them.”
—Johann Wolfgang Von Goethe (1749–1832)
“The hard truth is that what may be acceptable in elite culture may not be acceptable in mass culture, that tastes which pose only innocent ethical issues as the property of a minority become corrupting when they become more established. Taste is context, and the context has changed.”
—Susan Sontag (b. 1933)