Namespace: |
|
Content: |
empty, 4 attributes |
Defined: |
globally in ccapp.xsd, see XML source |
Includes: |
definitions of 4 attributes |
Used: |
at 1 location |
XML Representation Summary | |||
<... | |||
confidenceLevel | = |
xsd:double | |
globalOnly | = |
xsd:boolean : "false" | |
measure | = |
xsd:string | |
targetError | = |
xsd:double | |
/> |
By default, a constant number of observations, specified in the parameter file of the experiment, is obtained by simulation. Depending on the type of experiment, observations correspond to replications, or batches. The minimal number of observations is given by the minReplications or minBatches attributes, in experiment parameters.
However, if a target relative error is given for a set of performance measures selected using sequentialSampling elements in experiment parameters, the number of observations becomes random: after a minimum of n 0 observations are obtained, the simulator computes the estimates for the selected measures and their associated confidence intervals. Let X be an estimator for one of the selected performance measures, and let the confidence interval on the true performance measure be [ [X - D, X + D] with confidence level set by the attribute confidenceLevel. For each checked performance measure, the relative error D / X must be smaller than or equal to the threshold E selected by the targetError attribute. If this condition is violated for at least one checked performance measure, a new target number of observations is determined by rounding n 0 * (D / (E * |X|))^2 to the nearest integer, and additional observations are collected. In other words, the sample size increases until the required precision is attained. The process is repeated until the target relative error is reached, or a maximal number of observations given in experiment parameters is reached.
The measure attribute is used to determine the type of performance measures on which sequential sampling is performed. By default, the procedure is applied on each performance measure of the selected type, and the number of additional observations is the maximum of the numbers obtained for each measure. However, if the globalOnly attribute is set to true, only the performance measure for all contact types, agent groups, and periods, is checked.
If d performance measures are considered by the sequential sampling, with individual confidence levels 1-a 1 , ..., 1-a d , the overall confidence level for all the estimates to have a relative error smaller than the targets is 1-a 1 *...*a d .
<xsd:complexType name="SequentialSamplingParams"> </xsd:attribute> </xsd:attribute> </xsd:attribute> </xsd:attribute> </xsd:complexType> |
Type: |
|
Use: |
required |
Defined: |
<xsd:attribute name="confidenceLevel" type="ssj:double01" use="required"> </xsd:attribute> |
Type: |
xsd:boolean |
Use: |
optional |
Default: |
"false" |
Defined: |
<xsd:attribute default="false" name="globalOnly" type="xsd:boolean"> </xsd:attribute> |
Type: |
xsd:string |
Use: |
required |
Defined: |
<xsd:attribute name="measure" type="xsd:string" use="required"> </xsd:attribute> |
Type: |
|
Use: |
required |
Defined: |
<xsd:attribute name="targetError" type="ssj:nonNegativeDouble" use="required"> </xsd:attribute> |
XML Schema documentation generated with DocFlex/XML (Kit) v1.6.8 DocFlex/XML (Kit) is a freeware edition of DocFlex/XML, which is a powerful template-driven documentation and report generator from any data stored in XML files. Based on an innovative technology developed by FILIGRIS WORKS, this new tool offers virtuoso data querying and formatting capabilities not found in anything else! Need to convert your XML data into a clear nice-looking documentation or reports? Web-ready hypertext HTML or printable MS Word / OpenOffice.org friendly RTF? DocFlex/XML may be a cheap, quick and effective solution exactly for this task! Have questions? Not sure how to use it? Just send us e-mail to contact@filigris.com and we are always happy to help you! See also our services at www.filigris.com |