In: Statistics and Probability
Benford's Law states that the first nonzero digits of numbers drawn at random from a large complex data file have the following probability distribution.†
First Nonzero Digit | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
Probability | 0.301 | 0.176 | 0.125 | 0.097 | 0.079 | 0.067 | 0.058 | 0.051 | 0.046 |
Suppose that n = 275 numerical entries were drawn at random from a large accounting file of a major corporation. The first nonzero digits were recorded for the sample.
First Nonzero Digit | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
Sample Frequency | 85 | 50 | 34 | 22 | 20 | 18 | 13 | 17 | 16 |
Use a 1% level of significance to test the claim that the distribution of first nonzero digits in this accounting file follows Benford's Law.
b) Find the value of the chi-square statistic for the sample.
(Round the expected frequencies to at least three decimal places.
Round the test statistic to three decimal places.)
Given information :
Digit | Frequency(Oi) | probability(Pi) |
1 | 85 | 0.301 |
2 | 50 | 0.176 |
3 | 34 | 0.125 |
4 | 22 | 0.097 |
5 | 20 | 0.079 |
6 | 18 | 0.067 |
7 | 13 | 0.058 |
8 | 17 | 0.051 |
9 | 16 | 0.046 |
Total | 275 | |
N | 275 |
We have to compute Expected frequencies
Ei = N * Pi
Digit | Frequency(Oi) | probability(Pi) | Ei=N*Pi |
1 | 85 | 0.301 | 82.775 |
2 | 50 | 0.176 | 48.400 |
3 | 34 | 0.125 | 34.375 |
4 | 22 | 0.097 | 26.675 |
5 | 20 | 0.079 | 21.725 |
6 | 18 | 0.067 | 18.425 |
7 | 13 | 0.058 | 15.950 |
8 | 17 | 0.051 | 14.025 |
9 | 16 | 0.046 | 12.650 |
Total | 275 | 275 | |
N | 275 |
Hypothesis :
Ho : The distribution of first nonzero digits in this accounting file follows Benford's Law.
V/s
H1 : The distribution of first nonzero digits in this accounting file doesn't follows Benford's Law.
Under Ho,
Test statistic is,
Oi | Ei | (Oi-Ei)^2/Ei | |
85 | 82.775 | 0.060 | |
50 | 48.4 | 0.053 | |
34 | 34.375 | 0.004 | |
22 | 26.675 | 0.819 | |
20 | 21.725 | 0.137 | |
18 | 18.425 | 0.010 | |
13 | 15.95 | 0.546 | |
17 | 14.025 | 0.631 | |
16 | 12.65 | 0.887 | |
Total | 275 | 3.147 |
DF = k-1 = 8
We find critical value using chi square table.
Here < at 0.01 level of significance and 8 df then we fail to reject Ho.
At 0.01 level of significance there is sufficient evidance to support the claim that the distribution of first nonzero digits in this accounting file follows Benford's Law.