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:

    “Writing a book I have found to be like building a house. A man forms a plan, and collects materials. He thinks he has enough to raise a large and stately edifice; but after he has arranged, compacted and polished, his work turns out to be a very small performance. The authour however like the builder, knows how much labour his work has cost him; and therefore estimates it at a higher rate than other people think it deserves,”
    —James Boswell (1740–1795)