Weighted Mean - Vector-valued Estimates

Vector-valued Estimates

The above generalizes easily to the case of taking the mean of vector-valued estimates. For example, estimates of position on a plane may have less certainty in one direction than another. As in the scalar case, the weighted mean of multiple estimates can provide a maximum likelihood estimate. We simply replace by the covariance matrix:

W_i = \Sigma_i^{-1}.

The weighted mean in this case is:

\bar{\mathbf{x}} = \left(\sum_{i=1}^n \Sigma_i^{-1}\right)^{-1}\left(\sum_{i=1}^n \Sigma_i^{-1} \mathbf{x}_i\right),

and the covariance of the weighted mean is:

\Sigma_{\bar{\mathbf{x}}} = \left(\sum_{i=1}^n \Sigma_i^{-1}\right)^{-1},

For example, consider the weighted mean of the point with high variance in the second component and with high variance in the first component. Then

then the weighted mean is:

which makes sense: the estimate is "compliant" in the second component and the estimate is compliant in the first component, so the weighted mean is nearly .

Read more about this topic:  Weighted Mean

Famous quotes containing the word estimates:

    “And, by the way, who estimates the value of the crop which nature yields in the still wilder fields unimproved by man? The crop of English hay is carefully weighed, the moisture calculated, the silicates and the potash; but in all dells and pond-holes in the woods and pastures and swamps grows a rich and various crop only unreaped by man.”
    —Henry David Thoreau (1817–1862)