Scribd
Upload
291 views2 pages

Lilliefors Test For The Exponential Distribution

The Lilliefors test is used to test if a data set comes from an exponential distribution. It calculates the maximum vertical distance (D) between the empirical distribution function (EDF) of the data and the exponential distribution function. The null hypothesis is rejected if D is greater than the critical value (Dn,α) for a given sample size n and significance level α. As an example, the test is applied to a data set of 10 observations and fails to reject the null hypothesis that the data comes from an exponential distribution, as the calculated D of 0.2321 is less than the critical value of 0.258.

Uploaded by

Hadia Azhar2558
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
291 views2 pages

Lilliefors Test For The Exponential Distribution

The Lilliefors test is used to test if a data set comes from an exponential distribution. It calculates the maximum vertical distance (D) between the empirical distribution function (EDF) of the data and the exponential distribution function. The null hypothesis is rejected if D is greater than the critical value (Dn,α) for a given sample size n and significance level α. As an example, the test is applied to a data set of 10 observations and fails to reject the null hypothesis that the data comes from an exponential distribution, as the calculated D of 0.2321 is less than the critical value of 0.258.

Uploaded by

Hadia Azhar2558
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Lilliefors Test for the Exponential Distribution

Objective: The objective of the Lilliefors test is used to test the null hypothesis that data come from a
normally distributed population.
Parametric Counterpart: The parametric counterpart of the Lilliefors test for normality is the
Anderson-Darling test.

Procedure:
Step 1
𝐻𝑜 : The random sample has the exponential distribution
𝐻1 : The distribution 𝑋 is not exponential
Step 2
Level of significance: 𝛼 = 0.05 (unless otherwise stated)
Step 3
Test Statistic: 𝐷 = Sup|𝑆(𝑥) − 𝐹0 ∗ (𝑥)|
𝑥
Here 𝐹0 ∗ (𝑥) is computed as 1 − 𝑒 −𝑥 ⁄𝑥̅ = 𝐹0 ∗ (𝑧) = 1 − 𝑒 −𝑧 , where 𝑥̅ is the sample mean and
𝑧 = 𝑥 ⁄𝑥̅ for each observed 𝑥.
Step 4
Calculations:
Step 5
Critical Region: Reject 𝐻𝑜 if 𝐷 > 𝐷𝑛,𝛼

Step 6
Conclusion:
Example 1
Test the null hypothesis that the data given below arose from a one parameter exponential
distribution. 1.5, 2.3, 4.2, 7.1, 10.4, 8.4, 9.3, 6.5, 2.5, 4.6.
Solution:
Step 1
𝐻𝑜 : The random sample has the exponential distribution
𝐻1 : The distribution 𝑋 is not exponential
Step 2
Level of significance: 𝛼 = 0.05
Step 3
Test Statistic: 𝐷 = Sup|𝑆(𝑥) − 𝐹0 ∗ (𝑥)|
𝑥
Here 𝐹0 ∗ (𝑥) is computed as 1 − 𝑒 −𝑥 ⁄𝑥̅ = 𝐹0 ∗ (𝑧) = 1 − 𝑒 −𝑧 , where 𝑥̅ is the sample mean and
𝑧 = 𝑥 ⁄𝑥̅ for each observed 𝑥.
Step 4
Calculations:𝑥̅ = 5.68 ⇒ 𝑧 = 𝑥 ⁄5.68
𝑋 𝑧 = 𝑥 ⁄5.68 𝐹0 ∗ (𝑥) = 𝐹0 ∗ (𝑧) 𝑓 𝑐. 𝑓. 𝑆(𝑥) |𝑆(𝑥) − 𝐹0 ∗ (𝑧)| |𝑆(𝑥 − 𝜖) − 𝐹0 ∗ (𝑧)|
1.5 0.26 0.2321 1 1 0.1 0.1321 0.2321−0=0.2321
2.3 0.40 0.3330 1 2 0.2 0.1330 0.3330−0.1=0.2330
2.5 0.44 0.3561 1 3 0.3 0.0561 0.3561−0.2=0.1561
4.2 0.74 0.5226 1 4 0.4 0.1226 0.5226−0.3=0.2226
4.6 0.81 0.5551 1 5 0.5 0.0551 0.5551−0.4=0.1551
6.5 1.14 0.6816 1 6 0.6 0.1816 0.6816−0.5=0.1816
7.1 1.25 0.7135 1 7 0.7 0.0135 0.7135−0.6=0.1135
8.4 1.48 0.7721 1 8 0.8 0.0279 0.7721−0.7=0.0721
9.3 1.64 0.8055 1 9 0.9 0.0945 0.8055−0.8=0.0055
10.4 1.83 0.8397 1 10 1.0 0.1603 0.8397−0.9=0.0603

𝐷 = 0.2321
Step 5
Critical Region: Reject 𝐻𝑜 if 𝐷 > 𝐷𝑛,𝛼 ⇒ 𝐷 > 𝐷10,0.05 ⇒ 𝐷 > 0.258

Step 6
Conclusion: Cannot reject Ho.

You might also like