Problems 6, 13, 16, 21, 23, 25, 27c from the Chapter 3 handout.
G&S section 1.1, problems 1-5, 10.
Model solutions: http://www.math.dartmouth.edu/~doyle/docs/60.2012/hw/m60hw1.pdf
G&S section 1.1: 14,15,16
G&S section 1.2: 12,13,16,17,18,19,28
G&S section 2.1: 5,6,7,9
Model solutions: http://www.math.dartmouth.edu/~doyle/docs/60.2012/hw/m60hw2.pdf
G&S section 2.2: 5,6,12,20
G&S section 3.1: 22,23,24
G&S section 3.2: 6,10,13,18,34
For Ex. 34, Change `a good estimate ...' to `a bad estimate'--as Jie Zhong has pointed out. This incorrect estimate is based on a misprint in Feller. The correct estimate is m = n log n + n log(1/log 2). Using your program, show that this is a good estimate.'
G&S section 4.1: 11,15,20,27,44-45 (with simulations), 55-60.
G&S section 4.2: 11,13.
G&S section 5.1: 6,16,21,28,42,45.
G&S section 5.2: 10,26,36,37
G&S section 6.1: 12,16a,19,20
G&S section 6.2: 3,15,17,18,23,24,25,26,27,29
G&S section 6.3: 7,12,13,17,18
Additional problems: Write Matlab code to simulate a Markov chain, given by an arbitrary transition matrix P. Using this code
NOTE: Printed copies of G&S have screwed-up numbering for the later problems in section 6.2. To fix things up, relabel problem 26 as 25(b), 27 as 26, 28 as 27, 29 as 28 and 30 as 29. The result should agree with the online version here:
http://math.dartmouth.edu/~prob/prob/prob.pdf
G&S section 7.1: 4,8 (Use the Matlab function conv to do convolutions.)
G&S section 7.2: 5,15
For all of the following problems, compute the exact answer in addition to using the CLT approximation. For the CLT computations, whenever appropriate use the so-called 1/2-correction, as in G&S Example 9.2. Note that the answer key does not reliably use (or omit) the 1/2-correction.
G&S section 9.1: 1,3,5,7
G&S section 9.2: 1,4