**Reading
List on Bayes Linear suggested by Michael Goldstein**

Goldstein
(1998) __Bayes Linear
Analysis__.
Encyclopaedia of Statistical Sciences. Wiley. This is something I've
used in the past to introduce the ideas to people - but for the
purposes of the working group it might be both too broad in covering
too many topics, while not deep enough in what it does cover.

P.S.
Craig, M. Goldstein, A.H. Seheult, J.A. Smith (1996).__ Bayes
linear strategies for matching hydrocarbon reservoir history__,
(with
discussion) in Bayesian Statistics 5, eds. J.M. Bernardo, et al.,
Oxford University Press, 69-95.

There
is far more detail on what we actually do in P.S. Craig, M.
Goldstein, A.H. Seheult, J.A. Smith (1997). __Pressure
matching for hydocarbon reservoirs: a case study in the use____ of
Bayes linear strategies for large computer experiments__
(with discussion) in Case Studies in Bayesian
Statistics,
vol. III, eds.
C. Gastonis et al. 37-93. Springer-Verlag.

These papers are both about `history matching' or calibration. Our forecasting approach is in:

P.
S. Craig, M. Goldstein, J. C. Rougier and A. H. Seheult (2001)
__Bayesian
Forecasting Using Large Computer
Models__, JASA 96, 717-729 (notice that we manage to
forecast without calibrating first, which makes everything much more
tractable ).

Two
current papers are:

M.
Goldstein and J.Rougier (2006), __Bayes Linear
Calibrated Prediction for Complex Systems__, Journal of
the
American Statistical Association/, forthcoming. This paper was
previously entitled "Calibrated Bayesian forecasting using large
computer simulators". It mixes calibration with prediction
while retaining
tractability even for very high dimensions - still no MCMC!

M.
Goldstein and J.Rougier (2006), __Reified
Bayesian Modelling and Inference for Physical Systems__.
Journal of Statistical Planning and Inference/, forthcoming as a
discussion
paper, subject to final approval. This is our `linking
models with reality paper' - with example.

and
one more:

P.S.
Craig, M. Goldstein, A.H. Seheult, J.A. __Smith __(1998),
__Constructing
partial prior specifications
for models of complex physical systems__. The
Statistician,
47, 37-53, which discusses
belief specification for these models.