commons-math/xdocs/userguide/stat.xml

93 lines
5.3 KiB
XML

<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="./xdoc.xsl"?>
<!-- $Revision: 1.5 $ $Date: 2003/11/15 18:38:16 $ -->
<document url="stat.html">
<properties>
<title>The Commons Math User Guide - Statistics</title>
<author email="phil@steitz.com">Phil Steitz</author>
</properties>
<body>
<section name="1 Statistics">
<subsection name="1.1 Overview" href="overview">
<p>This is yet to be written. Any contributions will be greatfully
accepted!</p>
</subsection>
<subsection name="1.2 Univariate statistics" href="univariate">
<p>This is yet to be written. Any contributions will be gratefully
accepted!</p>
</subsection>
<subsection name="1.3 Frequency distributions" href="frequency">
<p>This is yet to be written. Any contributions will be gratefully
accepted!</p>
</subsection>
<subsection name="1.4 Bivariate regression" href="regression">
<p>This is yet to be written. Any contributions will be gratefully
accepted!</p>
</subsection>
<subsection name="1.5 Statistical tests" href="tests">
<p>This is yet to be written. Any contributions will be gratefully
accepted!</p>
</subsection>
<subsection name="1.6 Distribution framework" href="distributions">
<p>
The distribution framework provides the means to compute probability density
function (PDF) probabilities and cumulative distribution function (CDF)
probabilities for common probability distributions. Along with the direct
computation of PDF and CDF probabilities, the framework also allows for the
computation of inverse PDF and inverse CDF values.
</p>
<p>
In order to use the distribution framework, first a distribution object must
be created. It is encouraged that all distribution object creation occurs via
the <code>org.apache.commons.math.stat.distribution.DistributionFactory</code>
class. <code>DistributionFactory</code> is a simple factory used to create all
of the distribution objects supported by Commons-Math. The typical usage of
<code>DistributionFactory</code> to create a distribution object would be:
</p>
<source>DistributionFactory factory = DistributionFactory.newInstance();
BinomialDistribution binomial = factory.createBinomialDistribution(10, .75);</source>
<p>
The distributions that can be instantiated via the <code>DistributionFactory</code>
are detailed below:
<table>
<tr><th>Distribution</th><th>Factory Method</th><th>Parameters</th></tr>
<tr><td>Binomial</td><td>createBinomialDistribution</td><td><div>Number of trials</div><div>Probability of success</div></td></tr>
<tr><td>Chi-Squared</td><td>createChiSquaredDistribution</td><td><div>Degrees of freedom</div></td></tr>
<tr><td>Exponential</td><td>createExponentialDistribution</td><td><div>Mean</div></td></tr>
<tr><td>F</td><td>createFDistribution</td><td><div>Numerator degrees of freedom</div><div>Denominator degrees of freedom</div></td></tr>
<tr><td>Gamma</td><td>createGammaDistribution</td><td><div>Alpha</div><div>Beta</div></td></tr>
<tr><td>Hypergeometric</td><td>createHypogeometricDistribution</td><td><div>Population size</div><div>Number of successes in population</div><div>Sample size</div></td></tr>
<tr><td>t</td><td>createTDistribution</td><td><div>Degrees of freedom</div></td></tr>
</table>
</p>
<p>
Using a distribution object, PDF and CDF probabilities are easily computed
using the <code>cummulativeProbability</code> methods. For a distribution <code>X</code>,
and a domain value, <code>x</code>, <code>cummulativeProbability</code> computes
<code>P(X &lt;= x)</code> (i.e. the lower tail probability of <code>X</code>).
</p>
<source>DistributionFactory factory = DistributionFactory.newInstance();
TDistribution t = factory.createBinomialDistribution(29);
double lowerTail = t.cummulativeProbability(-2.656); // P(T &lt;= -2.656)
double upperTail = 1.0 - t.cummulativeProbability(2.75); // P(T &gt;= 2.75)</source>
<p>
The inverse PDF and CDF values are just as easily computed using the
<code>inverseCummulativeProbability</code>methods. For a distribution <code>X</code>,
and a probability, <code>p</code>, <code>inverseCummulativeProbability</code>
computes the domain value <code>x</code>, such that:
<ul>
<li><code>P(X &lt;= x) = p</code>, for continuous distributions</li>
<li><code>P(X &lt;= x) &lt;= p</code>, for discrete distributions</li>
</ul>
Notice the different cases for continuous and discrete distributions. This is the result
of PDFs not being invertible functions. As such, for discrete distributions, an exact
domain value can not be returned. Only the "best" domain value. For Commons-Math, the "best"
domain value is determined by the largest domain value whose cummulative probability is
less-than or equal to the given probability.
</p>
</subsection>
</section>
</body>
</document>