Interferogram time-weighted averaging for subsidence monitoring

Hello to Everyone.
I am working on analyzing the subsidence between 2018 and 2019 on quite a big area covered by a minimum of 5 images and I would like to ask you for an opinion on time-weighted averaging of three interferogram products (to minimalize atmospheric decorrelation) with temp. baseline 360 or 372 days with 12 (24) days difference between each other. I would like to know if some normalization must be done - e.g. based on perp baseline, common high coherence point or any other. Therefore in my displacement (LOS) products are huge differences and I am not sure if the averaging in this case is very useful or not. Also, I would like to know if you have an experience of whether it’s better to average terrain-corrected products or not.

Thank you for any ideas or comments.