umontreal.iro.lecuyer.contactcenters.msk.stat
Class StatCallCenterStat
java.lang.Object
umontreal.iro.lecuyer.contactcenters.msk.stat.AbstractCallCenterStatProbes
umontreal.iro.lecuyer.contactcenters.msk.stat.StatCallCenterStat
- All Implemented Interfaces:
- CallCenterStatProbes
public class StatCallCenterStat
- extends AbstractCallCenterStatProbes
Represents a set of statistical probes containing other statistics as
observations. An object of this class is constructed from another
implementation of CallCenterStatProbes
, and a type of statistic to
collect. For each matrix of probes defined in the inner implementation,
a clone is made and stored into this object. When
addStat()
is called, each matrix of probes in the inner call center
statistical object is retrieved, an intermediate matrix of observations is
constructed, and the resulting matrix is added into the corresponding clone
stored into this object.
This class is used for stratified sampling and randomized Quasi-Monte Carlo
simulation as follows. Each time a macroreplication or stratum is simulated,
statistics from a stratum or a randomization of a point set are available in
an inner CallCenterStatProbes
implementation. This class can obtain
the averages, the variances, or the standard deviations from the probes, and
add them to other matrices of statistical probes. This results in averages of
averages, averages of variances, etc.
Method Summary |
void |
addStat()
Adds new statistics to the probes defined by this object. |
Methods inherited from class umontreal.iro.lecuyer.contactcenters.msk.stat.AbstractCallCenterStatProbes |
getAverage, getConfidenceInterval, getMatricesOfStatProbes, getMatrixOfFunctionOfMultipleMeansTallies, getMatrixOfStatProbes, getMatrixOfTallies, getMatrixOfTallyStores, getMax, getMin, getPerformanceMeasures, getVariance, getVarianceOfAverage, hasPerformanceMeasure, init, initPerformanceMeasures |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
StatCallCenterStat
public StatCallCenterStat(CallCenterStatProbes stat,
StatType statType,
boolean fmm)
- Constructs a new group of call center statistical probes taking the
observations from the inner set of probes stat, and collecting
the statistic statType. This constructor creates a matrix of
statistical probes for each performance measure defined in stat,
by using the clone method. However, if fmm is
false, no matrix of statistical probes is added for functions of
multiple means tallies.
- Parameters:
stat
- the input call center statistical probes.statType
- the type of statistic collected.fmm
- determines if functions of multiple means are processed.
StatCallCenterStat
public StatCallCenterStat(CallCenterStatProbes stat,
StatType statType,
boolean fmm,
boolean keepObs)
addStat
public void addStat()
- Adds new statistics to the probes defined by this object. Each matrix of
probes in the inner call center statistical object is retrieved, an
intermediate matrix of observations is constructed, and the resulting
matrix is added into the corresponding clone stored into this object.
The way the matrix of observations is constructed depends on the type of
input matrix of probes. For matrices of tallies, element (i, j) of the
matrix of observations is given by computing a statistic (average,
variance, etc.) on the observations of the tally (i, j) of the inner
matrix of tallies. For matrices of functions of multiple means tallies, an
array of tallies corresponds to each element (i, j), and a statistic is
extracted from each element of this array of tallies. This results in a 3D
array compatible with
MatrixOfFunctionOfMultipleMeansTallies.add(double[][][])
.
To submit a bug or ask questions, send an e-mail to
Richard Simard.