Tuesday, October 12, 2010

Class Activity 2.1: Time per problem

General comment on CA2:

1) Problems 9 and 10 (the writeup) will be due Monday. For problem 9, compare the marginalized pdfs of alpha and beta for the case with and without consideration of sigma_[Fe/H]. When you get here Wednesday or thereafter, do not do the integrals analytically! Do your integration numerically.

2) The primary goals of this assignment are
  • to understand how to calculate the occurrence rate as a function of object properties (planet occurrence as a function of stellar metallicity).
  • present your results clearly and professionally
  • gain science writing experience
Keep these goals in mind as you work on the problem set.

3) Time spent debugging should decline monotonically throughout the semester. If you get discouraged after spending an hour chasing down a MATLAB or LaTeX error, take heart in that this is one fewer bug to chase down later as you pursue your research.

4) Ask for help, especially if you exceed any specific ETC below. Talk in person and via email to your classmates, Tim, or myself.

Comments and ETCs for homework problems:

Several students have requested an estimate for the amount of time one should spend on each problem. For Wednesday, problems 1-7 are "due," which means that you should be prepared to work on problems 8-10 in class with your partner. It looked like most teams finished through 4 in class, so here are my estimated times to completion (ETC) and notes for problems 5-7:

CA5) You may ignore my specific suggestion of assigning 60% completeness to half of the stars. The key to this problem is to investigate how detection sensitivity (completeness) factors into your estimate of f. You may assign Si to the stars in your sample as you see fit. Query your intuition beforehand; how do you expect S to affect your estimate of f? Then do the test.

This should take no longer than it took you to complete problem 4: ETC 20 minutes

CA6) I will send out the cross-listed data set thanks to Mike B. and Matt.

You should still
look up the FV05 paper on NASA ADS, and read the sections that describe their data and analysis so you have the necessary background for this problem (the intro would also be useful for context). ETC 30-40 minutes for reading the relevant sections

CA7) ETC 10 minutes of thought after examining Fig 5 in FV05.

Total ETC <= 1.2 hours

No comments:

Post a Comment