Six terms error (or full 2 port - 12 terms error) correctio

Hello,

I need to decide if to apply a "sophisticated" error correction for power ratios to a given system or to use something simple for power detection.
The problem is when apply the six terms error correction.
Hence, I woul like to ask if the values of those terms are complex numbers (magnitude and phase)?

The Agilen Application Note 1287-3 showes the formulas of the Two port error corection.
When setting the error terms at the other ports of the given models to zero, one can derive the formulas for one port error terms?

Is there a way or a method to "ignore" the phase of signal, but still to correctly measure the signal only by its magnitude (power detection)?

Please explain,

Thanks