Errata (2)
(for the future fourth printing)
Here are the changes between the second and third
printing. If you find any typo not listed here or there,
be sure to point them out to Christian Robert (xian@ceremade.dauphine.fr)
or George Casella (gcasella@stat.ufl.edu).
Your contribution will be acknowledged in the next printing. Thanks in
advance!

xi: Thanks added, plus a sentence in
memory of Richard Tweedie.

24: in Exercise 1.27,
in the second display d,q
should be d,h(q)

51: in the last display, exp^{a2/2}
should be e^{a2/2}

62: in Exercise 2.12c,
2logr^{2}~
U[0,1]
should be 2logU = r^{2}

137: The optimal size n_{i}^{*}
is proportional to the square root of the quantity on the rhs [As
stated in Fishman, 1996, page 300]

151: h_{A}
is the number of visits, not the average number of visits
[Thanks
to George Fishman!]

154: In Definition 4.4.8, where
A is Harris recurrent should read and A is Harris recurrent[Thanks
to George Fishman!]

171: In Definition 4.7.6, the
Markov chain (x_{n}) must be stationnary [Thanks to Christophe
Andrieu!]

181: In the last line,
c_{i}j should be c_{ij}

216: Quian should be Qian (with a similar
correction in the index)

281: In the autocorrelation formula,
cov
should be corr [Thanks to Christophe Andrieu!]

334: Correction of the reference
to Billio et al. (1999). The same p 446.
Last modified on Feb 25, 2002 9:55.