Let be a random variable with the property that there exists a non-negative function defined on such that
then is a continuous random variable and is its probability distribution function or PDF (). An example is is the probability that a quantum measurement returns a value between and .
We have some simple properties
and
The cumulant is a very useful construction
The cumulant is a very useful construction
Uniform random variable
Microcanonical distribution
In statistical mechanics the microcanonical distribution applies to a system of constant energy and assumes that all states, which are specification of coordinate and conjugate momentum for which are equally likely, so . For a harmonic oscillator and the number of states of energy between and is
Restrict attention to and , then for fixed there is only a single independent degree of freedom, either or , which is not uniformly distributed. Using
the number of states accessible per unit energy is
For the full phase space we have which increases with energy (what we call a normal system).
If we sample the -state of an oscillator we are most likely to get a value that the oscillator spends most of its time in,
Normal random variable
One of its most important properties is that if is normally distributed with parameters , then is normally distributed with parameters . This illustrates the use of the cumulant
Exponential and Gamma distributions
The exponential distribution has a parameter , and pdf and cumulant
The exponential distribution is memoryless, if represents a “waiting time”, until an event occurs, there is no dependence on how much time has already elapsed (i.e. ). The only continuous memory-less distributions are exponential.
The Gamma distribution also has parameters ,
An example is the canonical distribution for a normal system of particles, one whose entropy increases with energy as a power of the energy
,
Normal random variable cumulant
Begin with the pdf (probability distribution)
for which (let , )
is called the error function
Note that
The relation between the two is
The error function is included in the function libraries of most CAS systems such as REDUCE, Maple and Mathematica.
Example
is normally distributed with . Compute . We will get .
% In REDUCE...
on rounded;
load_package specfn;
% Check table on Ross p. 131
0.5*(1+erf(3.4/sqrt(2)));
%example 3a in Chapter 5 of Ross
0.5*(1+erf((5-3)/(3*sqrt(2))))-0.5*(1+erf((2-3)/(3*sqrt(2))));
% returns 0.3779
Cauchy distribution
The cumulative probability for the Cauchy distribution is simple
Beta distribution
is beta distributed if
For the record, the Beta distribution comes up in the “q-model” of the distribution of loads in a column of beads. Note that
CAS systems also have the beta and gamma functions, but the integral for the cumulative probability is something that must be numerically computed
load_package specfn;
procedure betacumulative(a,b,xmax);
begin
dx:=xmax/1000;
retval:=for n:=1:1000 sum (n*dx)^(a-1)*(1-n*dx)^(b-1)*dx;
return(retval/beta(a,b));
end;
betacumulative(3,3,1);
betacumulative(3,3,0.3);
Cauchy/Student’s distributions
Let and be independent, normally distributed. Let
(1)
Student’s -distribution may be a little less obvious. The -statistic can be written as in which is normally distributed being a sum/difference of means, and has an independent -variable distribution
so that is distributed as
(2)
This is a modified Cauchy distribution.
Functions of random variables
This is used to obtain the PDF of the function of a random variable from the PDF of , consider that
however this is also the probability the will be less than ;
Now just take the derivative using Newton’s law to get the PDF of (and thus proving Theorem 6.1);
This is the basis of the so-called inverse method for computing random deviates with some particular PDF.
Example
Let be a uniform deviate on the interval , meaning that . This is the type of built-in random number generator computer operating systems have. Find the PDF for a random number .
First we get our cumulant
and finally
if .
Example
If is uniformly distributed over , find the probabilty density function of .
Let
Example
Find the distribution of , where is a fixed constant and is uniformly distributed on . This arises in ballistics. if a projectile is launched at speed and angle from the horizontal, its range is .
Use the example of the inverse method, for ;
but we have , , and
and so
and so the ranges are distributed according to
The cumulative probability will be an inverse sine function (verify).
Example (Problem 5.16 in Ross)
In independent tosses of a coin the coin landed heads times. Is it reasonable to assume that the coin is fair?
What is the probability of landing within one of the mean for a normal distribution?
The probability of landing within two of the mean is
For coin tosses the probability of heads will be normally distributed about the mean with . This coin is turning up heads standard deviations beyond the mean, this is not a fair coin.
Example
If is uniformly distributed on find the pdf of .
Let