Univariate Statistics - MATLAB Recipes for Earth Sciences - page 66

v =
Columns 1 through 5
9.7383 10.3814 11.0245 11.6676 12.3107
Columns 6 through 8
12.9537 13.5968 14.2399

h e functions histogram and histcount provide numerous ways of binning
the data, of normalizing the data, and of displaying the histogram. In the
following sections, we ot en have to normalize the frequency distribution
such that the total number of observations sum(n) is one. h is can be achieved
by typing

h = histogram(corg,'Normalization','probability')
v = h.BinWidth * 0.5 + h.BinEdges(1:end-1)
n = h.Values

which yields

v =
Columns 1 through 5
9.5000 10.5000 11.5000 12.5000 13.5000
Column 6
14.5000

n =
Columns 1 through 5
0.0333 0.1000 0.2167 0.3333 0.2500
Column 6
0.0667

As an alternative way of plotting the data, the empirical cumulative
distribution function can be displayed using cdfplot (Fig. 3.1 b).

cdfplot(corg)

h e most important parameters describing the distribution are the measures
for central tendency and the dispersion about the average. h e most popular
measure for central tendency is the arithmetic mean.

Gallery
3.2

mean(corg)

ans =
12.3448

Since this measure is very susceptible to outliers, we can take the median as
an alternative measure of central tendency,

median(corg)