Hi, folks. PS10 solution sets will be available during Tuesday's review
session. At the end of the review, I'll put any solution sets that have
not been picked up in your GovGrads mailboxes in Littauer. Non-GovGrads
who cannot make the review session should email me about picking up their
solution set. Solution notes for PS10 are below and attached.
Good job on this problem set, and thanks again for all your hard work this
semester. Good luck on the exam!
Cheers,
Ryan
*****************************
Gov1000
Problem Set 10
Solution Notes
Problem 1c. Two common errors occurred on this part. First, many people
included the y-variable in their calculations of the robust center and
spread. See lecture slide 528, and note that the y-variable is not
included. Second, many people set the chi-squared cutoff point using 3
degrees of freedom. The degrees of freedom should be Rousseeuw and von
Zomeren's "p". "p" = 3 in the example of slide 528, and in R&vZ's
Stackloss example. However, p = 10 in the Problem 1c case.
Problem 2. If you have questions about this problem, check the Gov1000
email archives (a great resource for the final!). A clarifying message
about this problem is available at
http://lists.fas.harvard.edu/pipermail/gov1000-list/2004-December/001450.ht…
for example.
------------------------------------------
Ryan T. Moore ~ Government & Social Policy
Ph.D. Candidate ~ Harvard University
Homepage: http://www.people.fas.harvard.edu/~rtmoore/
Gov1000: http://www.courses.fas.harvard.edu/~gov1000/
Hi, all. The final will be available tomorrow after all review activities
are complete. It is due two weeks later, at 5pm, Tuesday, 18 January.
Since Alison and I have heard from relatively few people about tomorrow's
review, we'll plan to be available in CBRSS Room 31 tomorrow from 12noon -
3pm. Send us an email, or just stop by with any questions you have.
After tomorrow, mathematical and computational questions will be
off-limits. Unlike the problem sets, the final is a strictly individual
exercise.
Best of luck,
Ryan
------------------------------------------
Ryan T. Moore ~ Government & Social Policy
Ph.D. Candidate ~ Harvard University
Homepage: http://www.people.fas.harvard.edu/~rtmoore/
Gov1000: http://www.courses.fas.harvard.edu/~gov1000/
hello all, happy new year,
i was wondering if there was an easy way to get coefficients from R lm
output into a matrix, in such a way that i'd have only the numbers
themselves and not the name of the variable to which that coefficient
applies as well. that is, at the moment i can get the coefficients using:
coefficients(lm.out)[5]) for the 5th independent variable in the
regression, but this would give me elements in the matrix that look
something like:
aus$income
0.073
what i want is to be able to get the 0.073 without the 'label'. is there a
way of doing this without actually going through and doing it manually
myself?
thanks!
Lucy
Hi, everyone. I hope your winter break has been enjoyable. On Tuesday, 4
January, Alison and I will be holding individual or small-group review
sessions for the final exam. We'll be available most of the day in CBRSS.
We won't have specific agenda, so it's up to you to spend some time
generating your own questions. In order to make sure that everyone can
get their questions answered, we'd like everyone to reserve specific times
for review.
Please email me with
a) your preferred time(s) Tuesday,
b) how many people / who will be coming, and
c) the topic of your questions.
Obviously, we're flexible if you generate new questions between emailing
and showing up Tuesday, but the more advance notice we have, the better
prepared we can be.
See you all again soon,
Ryan
------------------------------------------
Ryan T. Moore ~ Government & Social Policy
Ph.D. Candidate ~ Harvard University
Homepage: http://www.people.fas.harvard.edu/~rtmoore/
Gov1000: http://www.courses.fas.harvard.edu/~gov1000/
