COUNTING STATISTICS

PRESENTED BY
VIKAS LAKHWANI
INTRODUCTION
 In any sample containing a large number of radioactive
atoms, some average number will disintegrate per unit
time
 But the exact number which disintegrate in any given
unit of time fluctuates around the average value.
 In counting applications, it is important to estimate this
fluctuation because it indicates the repeatability of
results of a measurement.
Statistical Parameters
 Mean :The mean is the arithmetic average of a set of
measurements.
 Median :The median is the middle value if there is an
odd number of measurements. It is the average of the
two middle values when an even number of
measurements has been taken.
 Mode :The mode is the most commonly observed
value.
Statistical Parameters(contd)
 Range : The range is the difference between the highest
and lowest measures.
 Deviation:The deviation is calculated by examining the
difference between a measurement and the mean of a set
of measurements.
 Variance:The variance is the sum of the squared
deviations of all measurements divided by the number of
measurements minus one.
 Standard Deviation: The standard deviation is defined
as the square root of the variance. Conceptually, the
standard deviation is the root mean square deviation of
the data from the mean.
Poisson Distribution
 It exists for situations in which the chances of one
specific outcome occurring are very small.
 Specifically, this distribution applies to situations in
which there is a large number of “participants” but the
probability of an event occurring to an individual
participant is very low.
 Each of these distributions is discrete, which means
that they take on only integer value.
Properties
 Defined only for non-negative integer values
 The probability of getting a certain result N when the true
value is m:
P(N;m)=e-mmN/N!
 Variance, 2, is defined as such that 68.3% of the
measurement results fall within   of the true value m.
 For Poisson distribution, 2 = m.
Gaussian Distribution
 Gaussian distribution is also called Normal Distribution.
 In the case of the Gaussian distribution, σ is unconstrained
and can take on any value independent of m. The formula
for computing the probability of observing the measurement
x for a specified m and σ is as follows:

 
P( x; m, )  1 / 2  2 e  x  m  / 2
2 2
Waveform
Gaussian Distribution(contd)
 For large value m, Poisson distribution can be
approximated by Gaussian distribution
 Gaussian distribution is defined for any value of x
(positive integer for Poisson)
 Variance 2 can have any value (mean value m for
Poisson).
 If there is additional random errors in nuclear counting
apart from the radionuclide decay, the results are
described by Gaussian distribution with variance,
2=m+(N)2
Percent Uncertainity
 It also known as the standard error
 It is a measure that compares the standard deviation (a
measure of the randomness or noise) with the mean .
 The percent uncertainty is calculated by :
S.D/N *100
As N increases, the standard deviation grows more slowly
and These deviations from the mean are perceived as
noise, and the effect of this noise depends on how it
affects the result.
Error propagation
 it is a way of combining two or more random signals
together to get the third.
 These equations assume that the errors are gaussian in
mature.
 It can be used to measure more than one quantity to get a
final result.
 It can also be used to combine several independent sources
of random errors on same measurement.
Rules
 Rule 1:
d(x-y)= d(x+y)= [(dx)ˆ2 + (dy)ˆ2] ˆ0.5
 Rule 2:
d(xy)/xy = d(x/y)/(x/y)= [(dx/x)ˆ2 + (dy/y)ˆ2] ˆ0.5
 Rule 3:
d(xˆ n)/(xˆn) = n(dx/x)
 Rule 4:
dF(x0) = |dF/dx| dx0 at x=0
Statistical Tests
 The Chi-Square Test
 The t-Test
 The Treatment of Outliers
 Linear Regression.
The chi-square test
 The χ2 (chi-square) test is used to test whether random
variations in a set of
measurements are in fact consistent with Poisson distribution.
 The formula for calculating χ2 is

2
  _
 N i N 
     ( n  1) SD
n 2
2
_ _
i 1
N N
The chi-square test(contd.)
 The calculated 2 value is then compared with the table
(degree of freedom df=n-1)
 P is the probability that the observed variation in a series of
n measurements from a Poisson distribution would equal or
exceed the calculated 2
 0.02<P<0.98 are acceptable with P=0.5 a perfect. P<0.01
the variation is too large and P>0.99 is too good to be true.
0.01<P<0.02 and 0.98<P<0.99 are suspicious.
The t-test
 T-test is used to test the significance of the difference
between the means from two sets of data that whether they
are in fact from the same Gaussian distribution.
 Different formulas are used for independent and paired data
sets
 Independent data are obtained from two different groups
whilst paired data have some kind of correlation between
the two measurements.
The t-test(contd.)
The formula for the independent data sets is (with mean values
X1 and X2, standard deviation SD1 andSD2, n=n1+n2-2 and
df=n)

X X 
t
1 2

 SD  SD 
2
1
2
2

n
The t-test(contd)
 The formula for the paired data set is
(with the difference average, Standard
deviation of the pair differences SD,
number of paired measurements n and
df=n-1)

X X 
t
1 2

 SD  SD 
2
1
2
2

n
The t-test(contd)
 The formula for the paired data set is
(with the difference average, Standard
deviation of the pair differences SD,
number of paired measurements n and
df=n-1)
_

n2||
t
SD
The t-test(contd.)
 The calculated t (from independent and paired data) is then
compared with the table.
 The probability of the data sets are in fact from the same
Gaussian distribution is less than the tabled probability if
the calculated t is larger than the tabled value. Less than 5%
is considered to be different
REFRENCES
 INME - Principles of Radiation Physics
Chapter 14 .pdf
 http://radiographics.rsna.org/content/19/3/765.f
ull
 wikipedia