Homework_2 - the following(i Say whether it is a Bayesian...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
PubH 8400-01, Richly-parameterized models, Spring 2010 Homework #2, assigned Thu, Feb 4 1. Finish deriving the degrees of freedom in the fitted values of the balanced one-way random effects model (in the lecture transparencies on page IB3/3). 2. Fit two different penalized splines to the global-mean surface temperature data, as follows. (a) Use a Bayesian or conventional approach, or do one of each. (b) Use one or more of these software approaches: (i) Write your own code in R to implement what we've done in class (ii) Use the SAS macros in RWC (iii) Use SemiPar in R (iv) Use WinBUGS. DO NOT USE OTHER APPROACHES unless you ask me first and I say it's OK. (c) Use any basis and number of knots, whatever you find interesting. (d) Report your work explicitly. For each distinct penalized spline fit, your report should include
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: the following: (i) Say whether it is a Bayesian or conventional analysis. (ii) Say which software you used; include any code you've written as an appendix. (iii) Describe the basis you used, including -- the number of knots and how you placed them (do not say "I used the program's default"; say what that default is) -- the type of basis functions (broken-stick linear, radial, etc.). (iv) Describe the penalty (probably simple, but describe it anyway). (v) Give a plot showing the data and the fit. You might want to put both fits on the same plot to compare them, particularly if you include confidence/posterior intervals. Write this in English prose with enough connecting sentences so I can follow it. I want you to practice presenting material, which includes writing. Due Thu Feb 11...
View Full Document

This note was uploaded on 11/21/2011 for the course PUBH 8400 taught by Professor Jimhodges during the Spring '10 term at Minnesota.

Ask a homework question - tutors are online