Feb. 16, 2012

This was a day filled with meetings... We had our group lunch with Tom Loredo with interesting discussion of stats project the students are working on. One of the ideas that came  up during the meeting was to think about the ways to publish the full likelihood information with the data. This information could be published as part of on-line data available in ADS.  However, how could we  represent such information and how to make sure that people would do it? We may need to come up with some examples.

I had a discussion with Alyssa and Vinay on addition of astrostatistics course to the astronomy graduate students program at HU. Vinay and I wrote a short bullet summary of possibly syllabus. After talking with Alyssa we came up  with some options that she will discuss at the curriculum review meeting. Important timing as this is apparently the course revision which it is going to affect the next decade. Hopefully we'll get some astrostat there. The students do not know what they are missing now, so it is hard to know how many of them would be interested in such a course.

I stopped by the SolStat workshop this afternoon to see Alex describing the conceptual difference between the likelihood and a posterior probability. Important distinction that requires shift in the ways one think about the data and probabilities.

P(theta | Y) = P ( Y| theta) * P(theta) / P(Y)

where for parameters theta P( Y | theta) represents the likelihood of observing data Y... while
P(theta | Y) represents the full posterior probability of the theta given that we observed Y.
P(theta) represents the prior.
Alex talk made me think about it again...


Attended a great CfA colloquium today on the spiral arms in the Galaxy! One of the best talks I've attended recently.  The observations are made in CO and given the velocity shifts one gets a 3D data that can provide the information about the molecular clouds location in the Milky Way. The tricky observations of the far side of the galaxy and modeling of the data shows that there is a bar at 3kpc and also a far arm, a possible extension to the Centaurus arm on the far side. Great talk and the data!

Malgosia has finished several plots of the lightcurves overplotted with the rms, and rms histograms. Now it is time for interpretation and possible connections to the supermassive BH.... fun work which requires some thinking.

Yesterday... I talked with Tracy during lunch. Interesting idea about the RM and the X-ray clusters of galaxies - the cool cores with radio sources display higher RM.

Feb 13, 2012

Working on calculating uncertainty on the flux. Sherpa functions which use simulations to access the uncertainty are not fully developed and in many cases the simulation halts. I have been trying to solve some of the issues in my script, but as of today my final table is still missing the numbers. I plan to try the new pyblocxs runs and see if I can get the set of parameters to use.

Feb.11, 2012

Last week was not as productive as I hoped for. However,  I managed to organize the first discussion on strategies for Chandra jets proposal.  Need to summarize and circulate notes today... The plan is to focus on the high z science and include the whole sample, but revise strategies to see if there is a more efficient way to accomplish the science goals. Have to work on the science this weekend.

Malgosia started writing up the variability case from Bozena's visit. We still need to understand issues with statistical uncertainty on measured excess variance. Including the standard equation seems to overestimate the error. I was considering the simulations to try to understand the behaviour of this uncertainty.

Giulia is working with the data and will try to use Brandon's radio to gamma-ray relation to place the points on the graph.  She is going to present the result of this work in the workshop early March, so we do have to have the results soon.

Still wondering about the HST proposal.. If I get the science case together we might be able to do it before Feb.24 deadline.

Feb.7, 2012

Started thinking about Chandra proposals this week. Met with Malgosia and Giulia yesterday to do some brainstorming. Malgosia will focus on the new sample of heavily absorbed AGNs in nearby galaxies,

while Giulia will continue her studies of the Fermi sources. The two projects are quite interesting and hopefully we'll get the observing time.

There is also Magda proposal on the new X-ray cluster that I should start...

The main topic of the discussion this week is a large project of high z radio sources with the jets activity. 

We did not get the time last year, but we need to try.

Feb.1, 2012

Bozena is visiting this week and we started analysis of X-ray variability in quasars. Basically downloaded the data and started analysis of the lightcurves. Malgosia is creating the lightcurves and cleaning it.
I just wrote a few lines of python to calculate rms histograms. We need to have the analysis done for our initial discussion today.

Andreas and Vinay initiated a discussion on image analysis yesterday. The old issues of detection of faint extended features, with not clearly defined borders comes back again. This will be some long- term work
and we decided to start involving students. In meantime I should just go back to the data and application of LIRA to my quasar data, but not until after the proposal deadlines on March 15.

Proposals... Started already thinking about project, but the real work starts next week.