Bit more confusing than i hoped

Partly I have not accomplished a lot this week because of the fickle nature of my program. When I first got the griding working, it looked perfect. Then I went to sleep and woke up the next day to start applying it.

For clusters with larger photometric errors, younger ages were always more favorable according to an exponential like distribution. The way I factor in the binary distribution allows for the program to think a red giant is a really noisy main sequence star. The blue error and visual error are uncorrelated in my program. For an actual binary star, it would increase in magnitude for both filters, not just one. A binary companion would not provide a blue magnitude boost while not emitting any visual light.

Thinking about it now, I should actually take that correlation into account when computing probabilities somehow. Might be tricky though. I would have to calculate the most likely q for each star, then compute the probability for each star. This would slow things down a lot. I would get some useful binary statistics out of it though.

From these problems, I've decided to run both methods on each cluster and use the one which seems more reasonable.

To control the Cross Entropy/ Bootstrapping Method, I am going to take prior information more into account. If part of the distribution is not a remotely conceivable visual fit, I will eliminate it. Also, sometimes a point does not seem grouped in the overall distribution and appears as some background noise. So to eliminate that I am discrediting all ages that only have one hit.

I am also going to start taking prior information into account when calculating the mass probability distribution. I know the star is a red giant and not a really noisy main sequence star.

The tech people finally figured out why I didn't have access to gio (separate account), and as gio is many times faster I can progress much more rapidly now. Hopefully. Always only hopefully.

I feel like this project is chasing me in circles. But I guess I don't know what is the best way until I try many possible ways. There is no simple solution that I can attain on the first try. There is no reason to have expected this to be easily accomplished. These facts are things I conceptually heard before but never truly understood as much as I do now.

PS- If I don't get this all done in a week I'll finish it up at home.