clarifications

This commit is contained in:
hyginn 2017-10-12 15:14:46 -04:00
parent 98cd17bfe0
commit bd79626a3d

View File

@ -58,26 +58,26 @@
# Let's get a few facts about probability distributions out of the way:
# The "support" of a probability distribution is the range of values that have a
# non-zero probability. The "domain" of a probability distribution is the range
# of probabilities that the distribution can take over its support. Think of
# this as the ranges on the x- and y-axis respectively. Thus the distribution
# The "support" of a probability distribution is the range of outcomes that have
# a non-zero probability. The "domain" of a probability distribution is the
# range of probabilities that the distribution can take over its support. Think
# of this as the ranges on the x- and y-axis respectively. Thus the distribution
# can be written as p = f(x).
# The integral over a probability distribution is always 1. This means: the
# distribution reflects the situation that an event does occur, any event, but
# there is not no event.
# there is not "no event".
# R's inbuilt probability distributions always come in four flavours:
# d... for "density": this is the probability density function (p. d. f),
# R's inbuilt probability functions always come in four flavours:
# d... for "density": this is the probability density function (p.d.f.),
# the value of f(x) at x.
# p... for "probability": this is the cumulative distribution function
# (c. d. f.). It is 0 at the left edge of the support, and 1 at
# (c.d.f.). It is 0 at the left edge of the support, and 1 at
# the right edge.
# q... for "quantile": The quantile function return the x value at which p...
# q... for "quantile": The quantile function returns the x value at which p...
# takes a requested value.
# r... for "random": produces random numbers that are distributed according
# to the p. d. f.
# to the p.d.f.
# To illustrate with the "Normal Distribution" (Gaussian distribution):