On p. 10 of the notes, we are told that: ----------- The 2 x 2 matrix: 2 -2 -2 2 has no saddle points. ----------- Now my question is, don't all 2 x 2 matrices have NO saddle points? If this isn't true, then what is an example of a 2 x 2 matrix with a saddle point please? Thanks in advance
If the 2 X 2 matrix is as shown below 1 9 7 8 in that case it has a saddle point given by 7 , which is max in the column but lowest in the row. So i think it is possible to have a saddle point for 2X2 matrix. Thanx Have you understood the Bayes criterion....please let me know?
thanks for that. I guess I didn't think a 2x2 matrix had a saddle point because saddle points seem to make most sense in a 3x3 matrix... a point which is the intersection of a minimum of one curve and a maximum of another curve. And you can't have a curve until you have at least 3 data points (ie. 3 x 3 matrix). still working on bayes criterion. for some reason i can't get past the first few chapters of this course!
