SlideShare a Scribd company logo
1 of 33
AB TESTING
INTRODUCTION TO AB TEST
This section contains theoretical information
about the AB test.
002
IN OTHER WORDS: INDEPENDENT TWO SAMPLE T TEST
It is used when it is desired to make a comparison between the
mean of two groups. In other words, if there are two main
population parameters whose true value is unknown, the
difference between these averages is to be examined and the
comparison is made over the samples obtained for this purpose.
3
WHAT ARE THE AB TEST IMPLEMENTATION STEPS?
◉ The hypothesis is established.
◉ The normality assumption is checked.
◉ The homogeneity of variance assumption is checked.
◉ If the assumptions are met, the parametric test is applied, if
not, the nonparametric test is applied.
4
LET’S START WITH THE ESTABLISHMENT OF THE HYPOTHESIS
FIRST!
◉ H0: µ1 = µ2
H1: µ1 ≠ µ2
5
◉ H0: µ1 >= µ2
H1: µ1 < µ2
◉ H0: µ1 <= µ2
H1: µ1 > µ2
The hypotheses of the AB test are as follows:
“
◉ H0: There is no statistically
significant difference between the
control group and the test group.
◉ H1: There is statistically significant
difference between the control
group and the test group.
6
WHAT DO THESE HYPOTHESIS DO?
Hypothesis testing is a statistical method used to test a belief.
Hypotheses are divided into two types among themselves: H0
and H1. The H0 hypothesis is called the 'hypothesis 0' and is
considered constant. As the 'alternative hypothesis', the H1
hypothesis comes into play.
7
“
H0 is rejected if the p-value is less
than 0.05. Otherwise it will not be
rejected.
8
THEN, WHAT IS P-VALUE?
In statistics, the p-value is a function used to measure how
extreme the observed sample results are based on a statistical
model.
If the model (hypothesis 0) is assumed to be true, the p-value is
the probability of obtaining values equal to or exceeding the
tested value.
9
“
If the p-value is less than the assumed
(acceptable) error amount α = 0.05,
the corresponding H0 hypothesis is
rejected.
10
LET'S CONTINUE WITH THE ASSUMPTION OF NORMALITY!
Shapiro-Wilks test is used for the assumption of normality and
it is decided whether the distribution is normal according to
the p-value.
11
“
◉ H0: There is no statistically significant
difference between the sample
distribution and the theoretical normal
distribution.
◉ H1: There is a statistically significant
difference between the sample
distribution and the theoretical normal
distribution.
12
NOW HOMOGENEOUS OF VARIANCE!
Levene test is used to examine the homogeneity of the
variances and it is decided whether the variances are
homogeneously distributed according to the p-value.
13
“
◉ H0: Variances are homogeneous.
◉ H1: Variances are not
homogeneous.
14
AND NOW AB TEST IS AVAILABLE!
The AB test can be applied to samples that pass the normality
assumption and variance homogeneity tests.
15
“
◉ ttest_ind test is applied when the
assumptions are satisfied.
◉ mannwhitneyu nonparametric test
is applied if the assumptions are
not satisfied.
16
01
-----
UNDERSTAND THE WORK
Sharing of the problem
02
-----
UNDERSTAND THE DATA
Basic information about
the dataset
03
-----
AB TESTING
Application of AB test after
assumption tests
17
UNDERSTAND THE WORK
0118
PROBLEM
abc.veribilimiokulu.com has the X suggestion system and they
received an offer to switch to the Y suggestion system.
They have decided to test the Y recommendation system and
want to do an AB test to see if it has more payoffs.
19
UNDERSTAND THE DATA
0220
THE STORY OF DATASET
The data set contains information on clicks and purchases
separately for abc.veribilimiokulu.com's X recommendation
system and Y recommendation system.
21
VARIABLES
22
VARIABLE NAME DESCRIPTION VARIABLE TYPE
Impression Ad views Numeric
Clicks Indicates the number
of clicks on the
displayed ad.
Numeric
Purchase Indicates the number
of products purchased
after the ads clicked.
Numeric
Earning Earnings after
purchased products
Numeric
AB TESTING
0323
NORMALITY
24
NORMALITY
p-value for the control group is greater than 0.05. ‘’H0: There is
no statistically significant difference between the sample
distribution and the theoretical normal distribution.’’ hypothesis
is not rejected. Distribution is normal.
25
NORMALITY
26
NORMALITY
p-value for the test group is greater than 0.05. ‘’H0: There is no
statistically significant difference between the sample
distribution and the theoretical normal distribution.’’ hypothesis
is not rejected. Distribution is normal.
27
HOMOGENCY OF VARIANCE
28
HOMOGENCY OF VARIANCE
p-value is greater than 0.05. ‘’ H0: Variances are homogeneous. ’’
hypothesis is not rejected. Parametric AB Test can be applied.
29
APPLICATION OF THE AB TEST
30
APPLICATION OF THE AB TEST
p-value is greater than 0.05. ‘’H0: There is no statistically
significant difference between the control group and the test
group.’’ hypothesis is not rejected.
31
INTERPRETATION OF AB TEST RESULT
There is no statistically significant difference between the
returns of these two recommendation systems. Therefore, it
would be unnecessary to allocate a budget to the new proposal
system as there is no difference. It can be continued with the
existing suggestion system.
32
THANKS!
Any questions?
You can find me at
baharzerenturk@gmail.com
linkedin.com/in/baharzerenturk1
BAHAR ZERENTÜRK
33

More Related Content

What's hot

Malimu statistical significance testing.
Malimu statistical significance testing.Malimu statistical significance testing.
Malimu statistical significance testing.
Miharbi Ignasm
 
Tests of significance by dr ali2003
Tests of significance by dr ali2003Tests of significance by dr ali2003
Tests of significance by dr ali2003
OSMAN ALI MD
 
Nonparametric tests
Nonparametric testsNonparametric tests
Nonparametric tests
Arun Kumar
 
Significance tests
Significance testsSignificance tests
Significance tests
Jinho Choi
 

What's hot (17)

Malimu statistical significance testing.
Malimu statistical significance testing.Malimu statistical significance testing.
Malimu statistical significance testing.
 
Significance test
Significance testSignificance test
Significance test
 
Tests of significance by dr ali2003
Tests of significance by dr ali2003Tests of significance by dr ali2003
Tests of significance by dr ali2003
 
Presentation chi-square test & Anova
Presentation   chi-square test & AnovaPresentation   chi-square test & Anova
Presentation chi-square test & Anova
 
Chi square test
Chi square testChi square test
Chi square test
 
Z-Test with Examples
Z-Test with ExamplesZ-Test with Examples
Z-Test with Examples
 
Nonparametric tests
Nonparametric testsNonparametric tests
Nonparametric tests
 
Z test
Z testZ test
Z test
 
Statistical inference: Statistical Power, ANOVA, and Post Hoc tests
Statistical inference: Statistical Power, ANOVA, and Post Hoc testsStatistical inference: Statistical Power, ANOVA, and Post Hoc tests
Statistical inference: Statistical Power, ANOVA, and Post Hoc tests
 
Fishers test
Fishers testFishers test
Fishers test
 
Significance tests
Significance testsSignificance tests
Significance tests
 
Chi square
Chi squareChi square
Chi square
 
Ch7 Analysis of Variance (ANOVA)
Ch7 Analysis of Variance (ANOVA)Ch7 Analysis of Variance (ANOVA)
Ch7 Analysis of Variance (ANOVA)
 
Parametric & non parametric
Parametric & non parametricParametric & non parametric
Parametric & non parametric
 
Test of significance
Test of significanceTest of significance
Test of significance
 
Statistical tests /certified fixed orthodontic courses by Indian dental academy
Statistical tests /certified fixed orthodontic courses by Indian dental academy Statistical tests /certified fixed orthodontic courses by Indian dental academy
Statistical tests /certified fixed orthodontic courses by Indian dental academy
 
Inferential statistics
Inferential statisticsInferential statistics
Inferential statistics
 

Similar to AB TESTING

Review Z Test Ci 1
Review Z Test Ci 1Review Z Test Ci 1
Review Z Test Ci 1
shoffma5
 
Chapter 20 and 21 combined testing hypotheses about proportions 2013
Chapter 20 and 21 combined testing hypotheses about proportions 2013Chapter 20 and 21 combined testing hypotheses about proportions 2013
Chapter 20 and 21 combined testing hypotheses about proportions 2013
calculistictt
 
Day-2_Presentation for SPSS parametric workshop.pptx
Day-2_Presentation for SPSS parametric workshop.pptxDay-2_Presentation for SPSS parametric workshop.pptx
Day-2_Presentation for SPSS parametric workshop.pptx
rjaisankar
 

Similar to AB TESTING (20)

20200519073328de6dca404c.pdfkshhjejhehdhd
20200519073328de6dca404c.pdfkshhjejhehdhd20200519073328de6dca404c.pdfkshhjejhehdhd
20200519073328de6dca404c.pdfkshhjejhehdhd
 
Hypothesis Tests in R Programming
Hypothesis Tests in R ProgrammingHypothesis Tests in R Programming
Hypothesis Tests in R Programming
 
Lund 2009
Lund 2009Lund 2009
Lund 2009
 
Review Z Test Ci 1
Review Z Test Ci 1Review Z Test Ci 1
Review Z Test Ci 1
 
Statistics for Lab Scientists
Statistics for Lab ScientistsStatistics for Lab Scientists
Statistics for Lab Scientists
 
Hypothesis Testing.pptx
Hypothesis Testing.pptxHypothesis Testing.pptx
Hypothesis Testing.pptx
 
Spss session 1 and 2
Spss session 1 and 2Spss session 1 and 2
Spss session 1 and 2
 
L hypo testing
L hypo testingL hypo testing
L hypo testing
 
What is the Independent Samples T Test Method of Analysis and How Can it Bene...
What is the Independent Samples T Test Method of Analysis and How Can it Bene...What is the Independent Samples T Test Method of Analysis and How Can it Bene...
What is the Independent Samples T Test Method of Analysis and How Can it Bene...
 
Testing of Hypothesis
Testing of Hypothesis Testing of Hypothesis
Testing of Hypothesis
 
Testing of Hypothesis combined with tests.pdf
Testing of Hypothesis combined with tests.pdfTesting of Hypothesis combined with tests.pdf
Testing of Hypothesis combined with tests.pdf
 
RESEARCH METHODS LESSON 3
RESEARCH METHODS LESSON 3RESEARCH METHODS LESSON 3
RESEARCH METHODS LESSON 3
 
Hypothesis
HypothesisHypothesis
Hypothesis
 
tps5e_Ch11_1.ppt
tps5e_Ch11_1.ppttps5e_Ch11_1.ppt
tps5e_Ch11_1.ppt
 
Chapter 20 and 21 combined testing hypotheses about proportions 2013
Chapter 20 and 21 combined testing hypotheses about proportions 2013Chapter 20 and 21 combined testing hypotheses about proportions 2013
Chapter 20 and 21 combined testing hypotheses about proportions 2013
 
83341 ch27 jacobsen
83341 ch27 jacobsen83341 ch27 jacobsen
83341 ch27 jacobsen
 
Business research method
Business research methodBusiness research method
Business research method
 
Day-2_Presentation for SPSS parametric workshop.pptx
Day-2_Presentation for SPSS parametric workshop.pptxDay-2_Presentation for SPSS parametric workshop.pptx
Day-2_Presentation for SPSS parametric workshop.pptx
 
Lecture2 hypothesis testing
Lecture2 hypothesis testingLecture2 hypothesis testing
Lecture2 hypothesis testing
 
8. testing of hypothesis for variable &amp; attribute data
8. testing of hypothesis for variable &amp; attribute  data8. testing of hypothesis for variable &amp; attribute  data
8. testing of hypothesis for variable &amp; attribute data
 

Recently uploaded

一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理
一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理
一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理
pyhepag
 
Fuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertaintyFuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertainty
RafigAliyev2
 
一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理
cyebo
 
一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理
cyebo
 
Abortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotec
Abortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotecAbortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotec
Abortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotec
Abortion pills in Riyadh +966572737505 get cytotec
 
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
pyhepag
 

Recently uploaded (20)

Artificial_General_Intelligence__storm_gen_article.pdf
Artificial_General_Intelligence__storm_gen_article.pdfArtificial_General_Intelligence__storm_gen_article.pdf
Artificial_General_Intelligence__storm_gen_article.pdf
 
Supply chain analytics to combat the effects of Ukraine-Russia-conflict
Supply chain analytics to combat the effects of Ukraine-Russia-conflictSupply chain analytics to combat the effects of Ukraine-Russia-conflict
Supply chain analytics to combat the effects of Ukraine-Russia-conflict
 
2024 Q1 Tableau User Group Leader Quarterly Call
2024 Q1 Tableau User Group Leader Quarterly Call2024 Q1 Tableau User Group Leader Quarterly Call
2024 Q1 Tableau User Group Leader Quarterly Call
 
Pre-ProductionImproveddsfjgndflghtgg.pptx
Pre-ProductionImproveddsfjgndflghtgg.pptxPre-ProductionImproveddsfjgndflghtgg.pptx
Pre-ProductionImproveddsfjgndflghtgg.pptx
 
MALL CUSTOMER SEGMENTATION USING K-MEANS CLUSTERING.pptx
MALL CUSTOMER SEGMENTATION USING K-MEANS CLUSTERING.pptxMALL CUSTOMER SEGMENTATION USING K-MEANS CLUSTERING.pptx
MALL CUSTOMER SEGMENTATION USING K-MEANS CLUSTERING.pptx
 
一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理
一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理
一比一原版加利福尼亚大学尔湾分校毕业证成绩单如何办理
 
Fuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertaintyFuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertainty
 
一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理
 
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPsWebinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
 
how can i exchange pi coins for others currency like Bitcoin
how can i exchange pi coins for others currency like Bitcoinhow can i exchange pi coins for others currency like Bitcoin
how can i exchange pi coins for others currency like Bitcoin
 
一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理
 
AI Imagen for data-storytelling Infographics.pdf
AI Imagen for data-storytelling Infographics.pdfAI Imagen for data-storytelling Infographics.pdf
AI Imagen for data-storytelling Infographics.pdf
 
Machine Learning for Accident Severity Prediction
Machine Learning for Accident Severity PredictionMachine Learning for Accident Severity Prediction
Machine Learning for Accident Severity Prediction
 
Abortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotec
Abortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotecAbortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotec
Abortion pills in Dammam Saudi Arabia// +966572737505 // buy cytotec
 
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
 
Slip-and-fall Injuries: Top Workers' Comp Claims
Slip-and-fall Injuries: Top Workers' Comp ClaimsSlip-and-fall Injuries: Top Workers' Comp Claims
Slip-and-fall Injuries: Top Workers' Comp Claims
 
How can I successfully sell my pi coins in Philippines?
How can I successfully sell my pi coins in Philippines?How can I successfully sell my pi coins in Philippines?
How can I successfully sell my pi coins in Philippines?
 
Atlantic Grupa Case Study (Mintec Data AI)
Atlantic Grupa Case Study (Mintec Data AI)Atlantic Grupa Case Study (Mintec Data AI)
Atlantic Grupa Case Study (Mintec Data AI)
 
How I opened a fake bank account and didn't go to prison
How I opened a fake bank account and didn't go to prisonHow I opened a fake bank account and didn't go to prison
How I opened a fake bank account and didn't go to prison
 
2024 Q2 Orange County (CA) Tableau User Group Meeting
2024 Q2 Orange County (CA) Tableau User Group Meeting2024 Q2 Orange County (CA) Tableau User Group Meeting
2024 Q2 Orange County (CA) Tableau User Group Meeting
 

AB TESTING

  • 2. INTRODUCTION TO AB TEST This section contains theoretical information about the AB test. 002
  • 3. IN OTHER WORDS: INDEPENDENT TWO SAMPLE T TEST It is used when it is desired to make a comparison between the mean of two groups. In other words, if there are two main population parameters whose true value is unknown, the difference between these averages is to be examined and the comparison is made over the samples obtained for this purpose. 3
  • 4. WHAT ARE THE AB TEST IMPLEMENTATION STEPS? ◉ The hypothesis is established. ◉ The normality assumption is checked. ◉ The homogeneity of variance assumption is checked. ◉ If the assumptions are met, the parametric test is applied, if not, the nonparametric test is applied. 4
  • 5. LET’S START WITH THE ESTABLISHMENT OF THE HYPOTHESIS FIRST! ◉ H0: µ1 = µ2 H1: µ1 ≠ µ2 5 ◉ H0: µ1 >= µ2 H1: µ1 < µ2 ◉ H0: µ1 <= µ2 H1: µ1 > µ2 The hypotheses of the AB test are as follows:
  • 6. “ ◉ H0: There is no statistically significant difference between the control group and the test group. ◉ H1: There is statistically significant difference between the control group and the test group. 6
  • 7. WHAT DO THESE HYPOTHESIS DO? Hypothesis testing is a statistical method used to test a belief. Hypotheses are divided into two types among themselves: H0 and H1. The H0 hypothesis is called the 'hypothesis 0' and is considered constant. As the 'alternative hypothesis', the H1 hypothesis comes into play. 7
  • 8. “ H0 is rejected if the p-value is less than 0.05. Otherwise it will not be rejected. 8
  • 9. THEN, WHAT IS P-VALUE? In statistics, the p-value is a function used to measure how extreme the observed sample results are based on a statistical model. If the model (hypothesis 0) is assumed to be true, the p-value is the probability of obtaining values equal to or exceeding the tested value. 9
  • 10. “ If the p-value is less than the assumed (acceptable) error amount α = 0.05, the corresponding H0 hypothesis is rejected. 10
  • 11. LET'S CONTINUE WITH THE ASSUMPTION OF NORMALITY! Shapiro-Wilks test is used for the assumption of normality and it is decided whether the distribution is normal according to the p-value. 11
  • 12. “ ◉ H0: There is no statistically significant difference between the sample distribution and the theoretical normal distribution. ◉ H1: There is a statistically significant difference between the sample distribution and the theoretical normal distribution. 12
  • 13. NOW HOMOGENEOUS OF VARIANCE! Levene test is used to examine the homogeneity of the variances and it is decided whether the variances are homogeneously distributed according to the p-value. 13
  • 14. “ ◉ H0: Variances are homogeneous. ◉ H1: Variances are not homogeneous. 14
  • 15. AND NOW AB TEST IS AVAILABLE! The AB test can be applied to samples that pass the normality assumption and variance homogeneity tests. 15
  • 16. “ ◉ ttest_ind test is applied when the assumptions are satisfied. ◉ mannwhitneyu nonparametric test is applied if the assumptions are not satisfied. 16
  • 17. 01 ----- UNDERSTAND THE WORK Sharing of the problem 02 ----- UNDERSTAND THE DATA Basic information about the dataset 03 ----- AB TESTING Application of AB test after assumption tests 17
  • 19. PROBLEM abc.veribilimiokulu.com has the X suggestion system and they received an offer to switch to the Y suggestion system. They have decided to test the Y recommendation system and want to do an AB test to see if it has more payoffs. 19
  • 21. THE STORY OF DATASET The data set contains information on clicks and purchases separately for abc.veribilimiokulu.com's X recommendation system and Y recommendation system. 21
  • 22. VARIABLES 22 VARIABLE NAME DESCRIPTION VARIABLE TYPE Impression Ad views Numeric Clicks Indicates the number of clicks on the displayed ad. Numeric Purchase Indicates the number of products purchased after the ads clicked. Numeric Earning Earnings after purchased products Numeric
  • 25. NORMALITY p-value for the control group is greater than 0.05. ‘’H0: There is no statistically significant difference between the sample distribution and the theoretical normal distribution.’’ hypothesis is not rejected. Distribution is normal. 25
  • 27. NORMALITY p-value for the test group is greater than 0.05. ‘’H0: There is no statistically significant difference between the sample distribution and the theoretical normal distribution.’’ hypothesis is not rejected. Distribution is normal. 27
  • 29. HOMOGENCY OF VARIANCE p-value is greater than 0.05. ‘’ H0: Variances are homogeneous. ’’ hypothesis is not rejected. Parametric AB Test can be applied. 29
  • 30. APPLICATION OF THE AB TEST 30
  • 31. APPLICATION OF THE AB TEST p-value is greater than 0.05. ‘’H0: There is no statistically significant difference between the control group and the test group.’’ hypothesis is not rejected. 31
  • 32. INTERPRETATION OF AB TEST RESULT There is no statistically significant difference between the returns of these two recommendation systems. Therefore, it would be unnecessary to allocate a budget to the new proposal system as there is no difference. It can be continued with the existing suggestion system. 32
  • 33. THANKS! Any questions? You can find me at baharzerenturk@gmail.com linkedin.com/in/baharzerenturk1 BAHAR ZERENTÜRK 33