Why should a part of the updated posterior variance using kalman filter have a negative sign?

11 Views Asked by At

enter image description here

Hi, I am learning the Kalman filter and am really confused at the estimated posterior variance. The conditional posterior variance has a negative term, but to me, it is supposed to have a positive sign. Could anyone explain why that (yellow colored part in the attached image) has a negative sign?

While deriving a variance, multiplying the same term with a negative sign naturally results in a positive sign.