Measurement Noise #2
Description
In the 15-year data set there is one pulsar that doesn't have an EFAC parameter for one of the backends in the noise files, but does have an EQUAD value. I'm not sure how this comes about, but it is not supported at the moment with the tempo 2 convention (which is also followed in enterprise) that uses both EFAC and EQUAD together.
One solution is to include adding extra flag parameters in the function to add measurement noise so that one can choose. I need to investigate how this pulsar is handled by enterprise in order to make a "consistent" solution. I suspect the right thing to do is to add a default EFAC value of 1 when EFAC is missing and the corresponding EQUAD value is nonzero for a particular backend.