encoding
latin1Fields and Methods
Methods:
rll{
getAsteriskTags -
getProbeAffinityFile -
}
Methods inherited from ProbeLevelModel:
calculateResidualSet, calculateWeights, fit, getAsteriskTags, getCalculateResidualsFunction, getChipEffectSet, getProbeAffinityFile, getResidualSet, getWeightsSet
Methods inherited from MultiArrayUnitModel:
getListOfPriors, setListOfPriors, validate
Methods inherited from UnitModel:
findUnitsTodo, getAsteriskTags, getFitSingleCellUnitFunction
Methods inherited from Model:
fit, getAlias, getAsteriskTags, getDataSet, getFullName, getName, getPath, getRootPath, getTags, setAlias, setTags
Methods inherited from Object:
asThis, getChecksum, $, $<-, [[, [[<-, as.character, attach, attachLocally, clearCache, clearLookupCache, clone, detach, equals, extend, finalize, gc, getEnvironment, getFieldModifier, getFieldModifiers, getFields, getInstantiationTime, getStaticInstance, hasField, hashCode, ll, load, objectSize, print, registerFinalizer, saveModel
For a single unit group, the affine model is:
$$y_{ik} = a + \theta_i \phi_k + \varepsilon_{ik}$$
where $a$ is an offset common to all probe signals,
$\theta_i$ are the chip effects for arrays $i=1,...,I$,
and $\phi_k$ are the probe affinities for probes $k=1,...,K$.
The $\varepsilon_{ik}$ are zero-mean noise with equal variance.
The model is constrained such that $\prod_k \phi_k = 1$.
Note that with the additional constraint $a=0$ (see arguments above),
the above model is very similar to MbeiPlm. The differences in
parameter estimates is due to difference is assumptions about the
error structure, which in turn affects how the model is estimated.