0

The inverse correlation matrix was calculated with the minverse function. The results show that in the diagonal the values are less than 1. how is this possible? the x1,x2 independent variables are correlated at about 0.5

stephen
  • 1
  • That is improbable, as you would expect the determinant and diagonal terms of the inverse matrix to be greater than $1$. For example $\begin{pmatrix} 1 & \frac12\ \frac12 & 1 \end{pmatrix}^{-1} = \begin{pmatrix} \frac43 & -\frac23\ -\frac23 & \frac43 \end{pmatrix}$. What was your correlation matrix? – Henry Mar 18 '22 at 00:19
  • Where is the function minverse? Language, package? Also show us your correlation matrix, and see https://stats.stackexchange.com/questions/120459/diagonal-elements-of-the-inverted-correlation-matrix – kjetil b halvorsen Mar 18 '22 at 00:29
  • henry thank you for the reply. hopefully you can get the below information it is column ozone solar wind temp ozone 1.0000000 0.5996416 0.2000837 -0.2395500 solar 0.5996416 1.0000000 0.0545718 -0.6240431 wind 0.2000837 0.0545718 1.0000000 -0.9028647 temp 0.2395500 -0.6240431 -0.9028647 1.0000000 – stephen Mar 18 '22 at 18:20

0 Answers0