23.02.2015 Views

Machine Learning - DISCo

Machine Learning - DISCo

Machine Learning - DISCo

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

section, if we were to repeat this experiment many times, each time drawing a<br />

different random sample Si of size n, we would expect to observe different values<br />

for the various errors,(h), depending on random differences in the makeup of<br />

the various Si. We say in such cases that errors, (h), the outcome of the ith such<br />

experiment, is a random variable. In general, one can think of a random variable<br />

as the name of an experiment with a random outcome. The value of the random<br />

variable is the observed outcome of the random experiment.<br />

Imagine that we were to run k such random experiments, measuring the random<br />

variables errors, (h), errors, (h) . . . errors, (h). Imagine further that we then<br />

plotted a histogram displaying the frequency with which we observed each possible<br />

error value. As we allowed k to grow, the histogram would approach the form<br />

of the distribution shown in Table 5.3. This table describes a particular probability<br />

distribution called the Binomial distribution.<br />

0.14<br />

0.12<br />

0.1<br />

0.08<br />

'F 0.06<br />

0.04<br />

0.02<br />

Binomial dishibution for n = 40, p =0.3<br />

0<br />

0 5 10 15 20 25 30 35 40<br />

A Binomial distribution gives the probability of observing r heads in a sample of n independent<br />

coin tosses, when the probability of heads on a single coin toss is p. It is defined by the probability<br />

function<br />

n!<br />

P(r) = pr(l - p)"-'<br />

r!(n - r)!<br />

If the random variable X follows a Binomial distribution, then:<br />

0 The probability Pr(X = r) that X will take on the value r is given by P(r)<br />

0 The expected, or mean value of X, E[X], is<br />

0 The variance of X, Var(X), is<br />

0 The standard deviation of X, ax, is<br />

Var (X) = np(1- p)<br />

For sufficiently large values of n the Binomial distribution is closely approximated by a Normal<br />

distribution (see Table 5.4) with the same mean and variance. Most statisticians recommend using<br />

the Normal approximation only when np(1- p) 2 5.<br />

TABLE 53<br />

The Binomial distribution.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!