Geoscience Reference
In-Depth Information
Fig. 7.12
Frequency distribution of subareas with
N
objects. In our example the subareas with
0, …, 5 objects are counted. h e histogram of the frequency distribution is displayed as a two-
dimensional histogram, using
histogram
.
Here, the midpoints of the histogram intervals
v
correspond to the
N
=0, …, 5
objects contained in the subareas. h e expected number of subareas
E
j
with
a certain number of objects
j
can be computed using
where
n
is the total number of objects and
T
is the number of subareas. For
j
=0,
j
! is taken to be 1. We compute the expected number of subareas, i.e., the
theoretical frequency distribution
n_exp
, using the equation shown above,
for i = 1 : 6
n_exp(i) = 49*exp(-100/49)*(100/49)^N(i)/factorial(N(i));
end
n_exp = sum(n_obs)*n_exp/sum(n_exp);
and display both the empirical and theoretical frequency distributions in a
single plot.
h1 = bar(v,n_obs);
hold on
h2 = bar(v,n_exp);
hold off
set(h1,'FaceColor','none','EdgeColor','r')
set(h2,'FaceColor','none','EdgeColor','b')
h e ˇ
2
-test is again used to compare the empirical and theoretical
distributions. h e test is performed at a
p
=95% signii cance level. Since the