Thesis Meeting
I had some preliminary results of running the pipeline on four different sources: 3C 111, Akn 120, IC 5063, and Mkn 348 I had tried using several different “minimum” time bins for extracting the spectra. Oddly enough, the change from 15 ks to 55 ks didn’t change the number of spectra by much for most of the sources (Akn 120 had 106 spectra for 55 ks bins, whereas it had 103 for 15 ks bins) — I think this is due to the count rate variability of the sources, though I may want to look closely at what the pipeline is doing just to make sure.
Unfortunately, there were some weird results with a couple of the sources. For example, in one of the Akn 120 spectra, the power-law index jumped up to 4 (which is not physically reasonable). The next step is going to be to run some well-studied (and published) data sets through the pipeline with a couple more models, and see if we can recover the published results. In addition, I will look through the pipeline results to find any spectra that are behaving badly, and re-analyze those by hand. We are agreed that I can’t possibly do all of the data by hand, but we are concerned that the pipeline is not giving physically reasonable results. So, we hope that most of the spectra are well-behaved, and that by re-analyzing the few that show odd behavior with the pipeline, I can reduce the number of spectra analyzed by had to a manageable number.