Suppose we have a [itex]mxn[/itex] matrix, where each row is an observation and each column is a variable. The [itex](i,j)[/itex]-element of its covariance matrix is [itex]\mathrm{E}\begin{bmatrix}(\vec{X_i} - \vec{\mu_i})^t*(\vec{X_j} - \vec{\mu_j})\end{bmatrix}[/itex], where [itex]\vec{X_i}[/itex] is the column vector corresponding to a variable (its elements are the observations) and [itex]\vec{\mu_i}[/itex] is the corresponding mean vector formed by one repeated element which is the mean value of the variable, calculated from its observations. Hence, the covariance matrix is a symmetric [itex]nxn[/itex] matrix.(adsbygoogle = window.adsbygoogle || []).push({});

Is the argument correct? Thank you for answering me.

Another question: could we just replace the "[itex]E[/itex] sign" with the division by the total number of observation, [itex]m[/itex]?

**Physics Forums | Science Articles, Homework Help, Discussion**

Join Physics Forums Today!

The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

# Dimensions of Covariance matrix (multiple observations)

**Physics Forums | Science Articles, Homework Help, Discussion**