math/doc/html/math_toolkit/stat_tut/overview/generic.html
2019-10-31 17:55:35 +00:00

259 lines
17 KiB
HTML

<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=US-ASCII">
<title>Generic operations common to all distributions are non-member functions</title>
<link rel="stylesheet" href="../../../math.css" type="text/css">
<meta name="generator" content="DocBook XSL Stylesheets V1.79.1">
<link rel="home" href="../../../index.html" title="Math Toolkit 2.11.0">
<link rel="up" href="../overview.html" title="Overview of Statistical Distributions">
<link rel="prev" href="objects.html" title="Distributions are Objects">
<link rel="next" href="complements.html" title="Complements are supported too - and when to use them">
</head>
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
<table cellpadding="2" width="100%"><tr>
<td valign="top"><img alt="Boost C++ Libraries" width="277" height="86" src="../../../../../../../boost.png"></td>
<td align="center"><a href="../../../../../../../index.html">Home</a></td>
<td align="center"><a href="../../../../../../../libs/libraries.htm">Libraries</a></td>
<td align="center"><a href="http://www.boost.org/users/people.html">People</a></td>
<td align="center"><a href="http://www.boost.org/users/faq.html">FAQ</a></td>
<td align="center"><a href="../../../../../../../more/index.htm">More</a></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="objects.html"><img src="../../../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../overview.html"><img src="../../../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../../../index.html"><img src="../../../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="complements.html"><img src="../../../../../../../doc/src/images/next.png" alt="Next"></a>
</div>
<div class="section">
<div class="titlepage"><div><div><h4 class="title">
<a name="math_toolkit.stat_tut.overview.generic"></a><a class="link" href="generic.html" title="Generic operations common to all distributions are non-member functions">Generic operations
common to all distributions are non-member functions</a>
</h4></div></div></div>
<p>
Want to calculate the PDF (Probability Density Function) of a distribution?
No problem, just use:
</p>
<pre class="programlisting"><span class="identifier">pdf</span><span class="special">(</span><span class="identifier">my_dist</span><span class="special">,</span> <span class="identifier">x</span><span class="special">);</span> <span class="comment">// Returns PDF (density) at point x of distribution my_dist.</span>
</pre>
<p>
Or how about the CDF (Cumulative Distribution Function):
</p>
<pre class="programlisting"><span class="identifier">cdf</span><span class="special">(</span><span class="identifier">my_dist</span><span class="special">,</span> <span class="identifier">x</span><span class="special">);</span> <span class="comment">// Returns CDF (integral from -infinity to point x)</span>
<span class="comment">// of distribution my_dist.</span>
</pre>
<p>
And quantiles are just the same:
</p>
<pre class="programlisting"><span class="identifier">quantile</span><span class="special">(</span><span class="identifier">my_dist</span><span class="special">,</span> <span class="identifier">p</span><span class="special">);</span> <span class="comment">// Returns the value of the random variable x</span>
<span class="comment">// such that cdf(my_dist, x) == p.</span>
</pre>
<p>
If you're wondering why these aren't member functions, it's to make the
library more easily extensible: if you want to add additional generic operations
- let's say the <span class="emphasis"><em>n'th moment</em></span> - then all you have to
do is add the appropriate non-member functions, overloaded for each implemented
distribution type.
</p>
<div class="tip"><table border="0" summary="Tip">
<tr>
<td rowspan="2" align="center" valign="top" width="25"><img alt="[Tip]" src="../../../../../../../doc/src/images/tip.png"></td>
<th align="left">Tip</th>
</tr>
<tr><td align="left" valign="top">
<p>
<span class="bold"><strong>Random numbers that approximate Quantiles of Distributions</strong></span>
</p>
<p>
If you want random numbers that are distributed in a specific way, for
example in a uniform, normal or triangular, see <a href="http://www.boost.org/libs/random/" target="_top">Boost.Random</a>.
</p>
<p>
Whilst in principal there's nothing to prevent you from using the quantile
function to convert a uniformly distributed random number to another
distribution, in practice there are much more efficient algorithms available
that are specific to random number generation.
</p>
</td></tr>
</table></div>
<p>
For example, the binomial distribution has two parameters: n (the number
of trials) and p (the probability of success on any one trial).
</p>
<p>
The <code class="computeroutput"><span class="identifier">binomial_distribution</span></code>
constructor therefore has two parameters:
</p>
<p>
<code class="computeroutput"><span class="identifier">binomial_distribution</span><span class="special">(</span><span class="identifier">RealType</span> <span class="identifier">n</span><span class="special">,</span> <span class="identifier">RealType</span>
<span class="identifier">p</span><span class="special">);</span></code>
</p>
<p>
For this distribution the <a href="http://en.wikipedia.org/wiki/Random_variate" target="_top">random
variate</a> is k: the number of successes observed. The probability
density/mass function (pdf) is therefore written as <span class="emphasis"><em>f(k; n, p)</em></span>.
</p>
<div class="note"><table border="0" summary="Note">
<tr>
<td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="../../../../../../../doc/src/images/note.png"></td>
<th align="left">Note</th>
</tr>
<tr><td align="left" valign="top">
<p>
<span class="bold"><strong>Random Variates and Distribution Parameters</strong></span>
</p>
<p>
The concept of a <a href="http://en.wikipedia.org/wiki/Random_variable" target="_top">random
variable</a> is closely linked to the term <a href="http://en.wikipedia.org/wiki/Random_variate" target="_top">random
variate</a>: a random variate is a particular value (outcome) of
a random variable. and <a href="http://en.wikipedia.org/wiki/Parameter" target="_top">distribution
parameters</a> are conventionally distinguished (for example in Wikipedia
and Wolfram MathWorld) by placing a semi-colon or vertical bar) <span class="emphasis"><em>after</em></span>
the <a href="http://en.wikipedia.org/wiki/Random_variable" target="_top">random
variable</a> (whose value you 'choose'), to separate the variate
from the parameter(s) that defines the shape of the distribution.
</p>
<p>
For example, the binomial distribution probability distribution function
(PDF) is written as <span class="serif_italic"><span class="emphasis"><em>f(k| n, p)</em></span>
= Pr(K = k|n, p) = </span> probability of observing k successes out
of n trials. K is the <a href="http://en.wikipedia.org/wiki/Random_variable" target="_top">random
variable</a>, k is the <a href="http://en.wikipedia.org/wiki/Random_variate" target="_top">random
variate</a>, the parameters are n (trials) and p (probability).
</p>
</td></tr>
</table></div>
<div class="note"><table border="0" summary="Note">
<tr>
<td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="../../../../../../../doc/src/images/note.png"></td>
<th align="left">Note</th>
</tr>
<tr><td align="left" valign="top"><p>
By convention, <a href="http://en.wikipedia.org/wiki/Random_variate" target="_top">random
variate</a> are lower case, usually k is integral, x if real, and
<a href="http://en.wikipedia.org/wiki/Random_variable" target="_top">random variable</a>
are upper case, K if integral, X if real. But this implementation treats
all as floating point values <code class="computeroutput"><span class="identifier">RealType</span></code>,
so if you really want an integral result, you must round: see note on
Discrete Probability Distributions below for details.
</p></td></tr>
</table></div>
<p>
As noted above the non-member function <code class="computeroutput"><span class="identifier">pdf</span></code>
has one parameter for the distribution object, and a second for the random
variate. So taking our binomial distribution example, we would write:
</p>
<p>
<code class="computeroutput"><span class="identifier">pdf</span><span class="special">(</span><span class="identifier">binomial_distribution</span><span class="special">&lt;</span><span class="identifier">RealType</span><span class="special">&gt;(</span><span class="identifier">n</span><span class="special">,</span> <span class="identifier">p</span><span class="special">),</span> <span class="identifier">k</span><span class="special">);</span></code>
</p>
<p>
The ranges of <a href="http://en.wikipedia.org/wiki/Random_variate" target="_top">random
variate</a> values that are permitted and are supported can be tested
by using two functions <code class="computeroutput"><span class="identifier">range</span></code>
and <code class="computeroutput"><span class="identifier">support</span></code>.
</p>
<p>
The distribution (effectively the <a href="http://en.wikipedia.org/wiki/Random_variate" target="_top">random
variate</a>) is said to be 'supported' over a range that is <a href="http://en.wikipedia.org/wiki/Probability_distribution" target="_top">"the smallest
closed set whose complement has probability zero"</a>. MathWorld
uses the word 'defined' for this range. Non-mathematicians might say it
means the 'interesting' smallest range of random variate x that has the
cdf going from zero to unity. Outside are uninteresting zones where the
pdf is zero, and the cdf zero or unity.
</p>
<p>
For most distributions, with probability distribution functions one might
describe as 'well-behaved', we have decided that it is most useful for
the supported range to <span class="bold"><strong>exclude</strong></span> random
variate values like exact zero <span class="bold"><strong>if the end point is
discontinuous</strong></span>. For example, the Weibull (scale 1, shape 1) distribution
smoothly heads for unity as the random variate x declines towards zero.
But at x = zero, the value of the pdf is suddenly exactly zero, by definition.
If you are plotting the PDF, or otherwise calculating, zero is not the
most useful value for the lower limit of supported, as we discovered. So
for this, and similar distributions, we have decided it is most numerically
useful to use the closest value to zero, min_value, for the limit of the
supported range. (The <code class="computeroutput"><span class="identifier">range</span></code>
remains from zero, so you will still get <code class="computeroutput"><span class="identifier">pdf</span><span class="special">(</span><span class="identifier">weibull</span><span class="special">,</span> <span class="number">0</span><span class="special">)</span>
<span class="special">==</span> <span class="number">0</span></code>).
(Exponential and gamma distributions have similarly discontinuous functions).
</p>
<p>
Mathematically, the functions may make sense with an (+ or -) infinite
value, but except for a few special cases (in the Normal and Cauchy distributions)
this implementation limits random variates to finite values from the <code class="computeroutput"><span class="identifier">max</span></code> to <code class="computeroutput"><span class="identifier">min</span></code>
for the <code class="computeroutput"><span class="identifier">RealType</span></code>. (See
<a class="link" href="../../sf_implementation.html#math_toolkit.sf_implementation.handling_of_floating_point_infin">Handling
of Floating-Point Infinity</a> for rationale).
</p>
<div class="note"><table border="0" summary="Note">
<tr>
<td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="../../../../../../../doc/src/images/note.png"></td>
<th align="left">Note</th>
</tr>
<tr><td align="left" valign="top">
<p>
<span class="bold"><strong>Discrete Probability Distributions</strong></span>
</p>
<p>
Note that the <a href="http://en.wikipedia.org/wiki/Discrete_probability_distribution" target="_top">discrete
distributions</a>, including the binomial, negative binomial, Poisson
&amp; Bernoulli, are all mathematically defined as discrete functions:
that is to say the functions <code class="computeroutput"><span class="identifier">cdf</span></code>
and <code class="computeroutput"><span class="identifier">pdf</span></code> are only defined
for integral values of the random variate.
</p>
<p>
However, because the method of calculation often uses continuous functions
it is convenient to treat them as if they were continuous functions,
and permit non-integral values of their parameters.
</p>
<p>
Users wanting to enforce a strict mathematical model may use <code class="computeroutput"><span class="identifier">floor</span></code> or <code class="computeroutput"><span class="identifier">ceil</span></code>
functions on the random variate prior to calling the distribution function.
</p>
<p>
The quantile functions for these distributions are hard to specify in
a manner that will satisfy everyone all of the time. The default behaviour
is to return an integer result, that has been rounded <span class="emphasis"><em>outwards</em></span>:
that is to say, lower quantiles - where the probablity is less than 0.5
are rounded down, while upper quantiles - where the probability is greater
than 0.5 - are rounded up. This behaviour ensures that if an X% quantile
is requested, then <span class="emphasis"><em>at least</em></span> the requested coverage
will be present in the central region, and <span class="emphasis"><em>no more than</em></span>
the requested coverage will be present in the tails.
</p>
<p>
This behaviour can be changed so that the quantile functions are rounded
differently, or return a real-valued result using <a class="link" href="../../pol_overview.html" title="Policy Overview">Policies</a>.
It is strongly recommended that you read the tutorial <a class="link" href="../../pol_tutorial/understand_dis_quant.html" title="Understanding Quantiles of Discrete Distributions">Understanding
Quantiles of Discrete Distributions</a> before using the quantile
function on a discrete distribtion. The <a class="link" href="../../pol_ref/discrete_quant_ref.html" title="Discrete Quantile Policies">reference
docs</a> describe how to change the rounding policy for these distributions.
</p>
<p>
For similar reasons continuous distributions with parameters like "degrees
of freedom" that might appear to be integral, are treated as real
values (and are promoted from integer to floating-point if necessary).
In this case however, there are a small number of situations where non-integral
degrees of freedom do have a genuine meaning.
</p>
</td></tr>
</table></div>
</div>
<table xmlns:rev="http://www.cs.rpi.edu/~gregod/boost/tools/doc/revision" width="100%"><tr>
<td align="left"></td>
<td align="right"><div class="copyright-footer">Copyright &#169; 2006-2019 Nikhar
Agrawal, Anton Bikineev, Paul A. Bristow, Marco Guazzone, Christopher Kormanyos,
Hubert Holin, Bruno Lalande, John Maddock, Jeremy Murphy, Matthew Pulver, Johan
R&#229;de, Gautam Sewani, Benjamin Sobotta, Nicholas Thompson, Thijs van den Berg,
Daryle Walker and Xiaogang Zhang<p>
Distributed under the Boost Software License, Version 1.0. (See accompanying
file LICENSE_1_0.txt or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>)
</p>
</div></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="objects.html"><img src="../../../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../overview.html"><img src="../../../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../../../index.html"><img src="../../../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="complements.html"><img src="../../../../../../../doc/src/images/next.png" alt="Next"></a>
</div>
</body>
</html>