etahat {calibrator}

Expectation of computer output

Description

Returns the apostiori expectation of the computer program at a particular point with a particular set of parameters, given the code output.

Usage

etahat(D1, D2, H1, y, E.theta, extractor, phi)

Arguments

D1

Matrix of code observation points and parameters

D2

Matrix of field observation points

H1

Basis functions

y

Code observations corresponding to rows of D1

E.theta

expectation wrt theta; see details

extractor

Extractor function

theta

Parameters

phi

Hyperparameters

Details

Argument E.theta is officially a function that, given x,y returns E_\theta\left(h_1(x,\theta)\right).

However, if supplied a non-function (this is tested by is.function() in the code), E.theta is interpreted as values of \theta to use. Recycling is carried out by function D1.fun()

Author(s)

Robin K. S. Hankin

References

M. C. Kennedy and A. O'Hagan 2001. Bayesian calibration of computer models. Journal of the Royal Statistical Society B, 63(3) pp425-464

M. C. Kennedy and A. O'Hagan 2001. Supplementary details on Bayesian calibration of computer models, Internal report, University of Sheffield. Available at http://www.tonyohagan.co.uk/academic/ps/calsup.ps

R. K. S. Hankin 2005. Introducing BACCO, an R bundle for Bayesian analysis of computer code output, Journal of Statistical Software, 14(16)

See Also

p.page4

Examples

data(toys)

etahat(D1=D1.toy, D2=D2.toy, H1=H1.toy, y=y.toy,
    E.theta=E.theta.toy, extractor=extractor.toy, phi=phi.toy)

# Now try giving E.theta=1:3, which will be interpreted as a value for theta:
etahat(D1=D1.toy, D2=D2.toy, H1=H1.toy, y=y.toy, E.theta=1:3,
     extractor=extractor.toy, phi=phi.toy)