Mike West
The Arts & Sciences Professor of Statistics & Decision Sciences
Department of Statistical Science
Duke University

Home
Research
Publications
Conferences
Students & Fellows
Videos
Software & data
Teaching
Contact & Directions

BEST Award
Stat@Duke

### Bayesian Forecasting and Dynamic Models Mike West and Jeff Harrison

Here are two paragraphs that were somehow lost from the 2nd printing at final stages, together with a rolling/updated list of typos spotted by readers over the years since publication. We will add any others that may be identified by diligent readers, with thanks.

Note that we use some basic TeX notation, e.g., "X_t" is "X" subscripted "t", and so forth

Chapter 1, page 31. The following paragraph should appear at the end of Chapter 1, p31:

The field continues to develop and flourish. As we approach the new millenium, we see exciting developments in new fields of application, and increasing sophistication in modelling developments and advanced computation. Interested readers might explore some recent studies in Aguilar and West (1998a,b), Aguilar, Huerta, Prado and West (1999), Cooper and Harrison (1997), and Prado, Krystal and West (1999), for example. Readers interested in contacting at least some of the more recently documented developments, publications and software, can explore the resources and links on the author web site indicated in the Preface.

Chapter 3

• p92, exercise 6, line -1: the statement C_t^{-1} \to 0 and is incorrect and should be whited out.

Chapter 10

• p362, Table 10.4, 2nd line under "Information"
\delta is missing from the T dof; it should read
(\theta_t|D_{t-1}) \sim T_{\delta n_{t-1}}[a_t,R_t]
• p362, Table 10.4, 2nd line under "Forecast:"
In the equation for Q_t the term S_{t-1} should be k_t S_{t-1}
• p364, in the final equation for n_t(-k) the term + \delta n_{t-k+1} should be +\delta n_t(-k+1)

Chapter 11

• p379 in line 4 of the statement of Lemma 11.1.
E[\theta_t] and V[\theta_t] should, of course, be E[\theta_t|D_{t-1}] and V[\theta_t|D_{t-1}]

Chapter 12

• p464 in equation (12.30), as well as line 8 on p468, and also equation (12.40) on p468, e_t(j_t, j_{t-1}) should be e_t(j_{t-1})
• p469 at the top - changes below and in this pdf version
• p_t^*(j_t) should be p_t^*(j_t , j_{t-1}) in both its definition in the displayed equation of line 4 and in the summation of line 5
• The summation of line 5 should be over index j_{t-1} not j_t
• Similarly, p_t^*(j_{t-1}) should be p_t^*(j_t , j_{t-1}) in the displayed equations of lines 7 and 9

Chapter 14

• p529, line -8 and -5.
A TeX typesetting error has led to the Gamma function \Gamma(*) being rendered as o(*) in the expressions for the Beta function, i.e., in normalizing constant terms for the beta density
• p559, exercise 5, part (b):
• E[Y_t|\mu_t] = \log(\lambda_t) should be E[Y_t|\mu_t] = \mu_t = \exp(\lambda_t)
• q_t=\dot\gamma(s_t) should be q_t=\dot\gamma(r_t)

Chapter 16

• p601, first bulleted item, second line.
The equation for E(\Phi) should be E(\Phi)= S^{-1} (n+q-1)/n

Chapter 17

• p642, line 4 in item (1)
There should be no /2 factor in the formula defining d*
• p651, the following paragraph should appear at the end of this chapter:
As an aside, note the more general Jordan forms for G matrices of non-observable models that appear in the proof of Theorem 5.2 (the requisite generalisations of the linear algebraic theory can be found, for example, in Theorem 8.5 of Nerig (1969).) In such cases any system matrix with one eigenvalue e of multiplicity n can be reduced to a form diag(J_{r_1}(e),...,J_{r_m}(e)) with r_1+...+r_m=n and where each J_{r_i}(e) matrix is a standard Jordan block. This might even be the diagonal case where each r_i=1 when G=eI. This completes the general theory but is of little practical interest.