17.01.2015 Views

GPS-X Technical Reference

GPS-X Technical Reference

GPS-X Technical Reference

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Optimizer 388<br />

In <strong>GPS</strong>-X it is assumed that the experimental errors are normally distributed random<br />

variables with a mean of zero. As a result, the likelihood function used in <strong>GPS</strong>-X is:<br />

Equation 14.6<br />

where:<br />

n<br />

m<br />

V i<br />

= number of experiments (i.e. observations)<br />

= number of measured response variables<br />

= variance-covariance matrix for the ith experiment<br />

| V i | = represents the determinant of V i<br />

ei<br />

<br />

= m x 1 residual vector that contains the differences between the<br />

measured values of the response variables and the values predicted by<br />

our mathematical process model.<br />

= the vector of parameters to be estimated in our mathematical process<br />

model.<br />

This expression is derived by substituting the residual vector, ei for the error vector in the<br />

multivariate normal probability density function (pdf). The error vector in the<br />

multivariate normal pdf contains the differences between the measured values of the<br />

response variables (the variables that we are fitting) and the true values. See Bard (1974)<br />

for details.<br />

<strong>GPS</strong>-X also assumes that the measurement errors are independent from observation to<br />

observation and from response variable to response variable. This means that the<br />

variance-covariance matrices found in Equation 14.6 are diagonal but not necessarily<br />

equal.<br />

<strong>GPS</strong>-X <strong>Technical</strong> <strong>Reference</strong>

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!