Scribd
Upload
72 views13 pages

Non Parametric Tests - A

Uploaded by

Roshan James
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
72 views13 pages

Non Parametric Tests - A

Uploaded by

Roshan James
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

GM103C DATA-ANALYTICS

10. NON-PARAMETRIC TESTS


1
NON-PARAMETRIC TESTS

Non-Parametric Tests
1. The one-sample runs test, a method for determining the randomness with which sampled items have been selected.
2. The sign test for paired data, where positive or negative signs are substituted for quantitative values.
3. A rank sum test, often called the Mann-Whitney U test, which can be used to determine whether two independent
samples have been drawn from the same population. It uses more information than the sign test. Equality of Two
Means.
4. Another rank sum test, the Kruskal-Wallis test, which generalizes the analysis of variance discussed to enable us to
dispense with the assumption that the populations are normally distributed. Equality of Several Means.
5. Rank correlation, a method for doing correlation analysis when the data are not available to use in numerical form,
but when information is sufficient to rank the data first, second, third, and so forth.
6. The Kolmogorov–Smirnov test, another method for determining the goodness of fit between an observed sample
and a theoretical probability distribution. Also Comparing two populations.
2
NON-PARAMETRIC TESTS

Advantages of Nonparametric Methods Non-Parametric Tests


 They do not require us to make the assumption that a population is distributed in the shape of a normal curve or
another specific shape.
 Generally, they are easier to do and to understand.
 Sometimes even formal ordering or ranking is not required.

Disadvantages of Nonparametric Methods


 They ignore a certain amount of information.
 They are often not as efficient or “sharp” as parametric tests.

3
4
NON-PARAMETRIC TESTS

Non-Parametric Tests
 The one-sample runs test, a method for determining the randomness with which sampled items have been selected.
 For testing whether the observations in a sample occur in a certain order or they occur in a random order.

   Step 1: Arrange the all observations in the order they are collected.
 Step II: Calculate the Median
 Step III: Mark + if the observation is above the Median and – if the observation is below the Median.
 Succession of values with same sign is called RUN.
 If the R values in between acceptance level, Accept Ho.

5
NON-PARAMETRIC TESTS

1. Test for Randomness in a series of Observations - One-sample runs test


Illustration 1
The director of an institute wanted to peruse the answer books of 20 students out of 250 students in the paper on Data
Analytics. The Registrar selected 20 answer books one by one from the entire lot of 250. The books had the marks
obtained by the students indicated on the first page. The director wanted to ascertain weather the Registrar had selected
the answer books in a random order.

58 61 78 72 69 65 79 75 80 81 82 79 58 77 71 73 74 79 81 82

6
58 61 78 72 69 65 79 75 80 81 82 79 58 77 71 73 74 79 81 82

7
8
NON-PARAMETRIC TESTS

2. Test for Specified Mean Or Median of a Population – The Signed Rank Test
 To investigate the significance of the difference between a population mean or median and a specified value of mean
or median, say

  
Illustration 2,
In the previous example, the Director further desired to have an idea of the average marks of the students. When he
enquired from the concerned Professor, he was informed that the Professor had not calculated the average but felt that
the mean would be 70, and the median would be 65. The Director wanted to test tis, and asked for a sample of 10
randomly selected answer book. (Significance Level 5%)

9
Marks (Xi) 55 58 63 78 72 69 66 79 75 80

10
11
REFERENCES

 Statistics for Managers Using Microsoft® Excel David M. Levine, David F. Stephan, Kathryn A. Szabat.
 Runs Test Table | Real Statistics Using Excel (real-statistics.com)
 Wilcoxon Signed-Ranks Table | Real Statistics Using Excel (real-statistics.com)

 How to select appropriate statistical tests?

https://www.youtube.com/watch?v=ZX4m6zav5To

12
THANK YOU
DR T SAKTHI NAGARAJ
ASSISTANT PROFESSOR, SCHOOL OF BUSINESS
AMRITA VISHWA VIDYAPEETHAM
COIMBATORE – 641112 13

CELL: 8610808385

You might also like