SlideShare ist ein Scribd-Unternehmen logo
1 von 10
Downloaden Sie, um offline zu lesen
Randomization Tests – unequal-N,
      unequal-σ problem




            AK Dhamija
Agenda
Assumptions of t, F tests
Randomization tests
Problems of Randomization Test
  Too liberal
  Too conservative
  Computationally Intensive
Solving the problems
  Resampling
  Gill’s algorithm
Assumptions of t, F tests

The two samples are each drawn from
normal distributions.

The two samples are drawn randomly from
their respective populations.

  RANDOMIZATION TESTS TACKLE THESE
  UNREALISTIC ASSUMPTIONS
Randomization tests
An Example Comparing t-Test and Randomization Test Results
    Two fertilizers (A and B) that are randomly applied to a type of sunflower seed.
    The maximum heights reached (in feet) are recorded after some time period.
    All Other Factors are constant

Null hypothesis : no difference between fertilizers A and B with respect to sunflower height.
Alternative hypothesis : fertilizer A is superior to fertilizer B on average with respect to sunflower height.

Sample      Fertilizer              Height (ft)
1           A                       9.9
2           B                       9.6
3           B                       9.7
4           B                       9.4
5           A                       10.1
6           B                       9.5
7           A                       9.9
8           B                       9.6         Total 462 (11 !/ 5! 6!) permutations
9           A                       9.5         5 of the 462 showed mean difference of 9.920 – 9.533 = 0.387
10          A                       10.2        p-value = 5/ 462 = 0.0108 => Reject H0 (t-test also rejects)
11          B                       9.4         = > fertilizer A outperforms fertilizer B
            So t-test provides reasonably good approximation to randomization test
Randomization tests
Randomization Tests do not consider normality, random sampling, equal variances, or
   other assumptions.

   The conclusion was based solely on the observed results, and the fact that the fertilizers
   were randomly assigned.

Why randomization tests then are not widely used, nor addressed in many statistical
  texts.

   The number of computations with larger sample sizes becomes astronomical
   With two samples, each of size 30, there are over 1.18 * 1017 possible permutations!

But randomization tests becomes sensitive to heteroscedasticity when the cells are
   unequal in size

   Approximate randomization Tests (selecting few combinations)
        Unstable – (statistics may vary)
        Unreplicable
Randomization tests
Full Randomization Test Problems (similar to t,F test)
    Too conservative if larger cells have larger variances (large effect is required for
    significance)
    Too liberal if smaller cells have larger variances (exaggerates the true difference)

                                          Variance Ratios
N    n1,n2 C(N,n1)    1:10     1:4        1:2       1:1        2:1       4:1        10:1
16   8,8       12,870 .0744    .0585      .0594     .045       .0616     .0464      .0656
20   8,14     125,970 .0312    .03        .0319     .058       .0921     .0984      .1152
24   8,16     735,471 .0156    .0158      .0181     .0468      .1222     .1304      .1618
28   8,20   3,108,105 .0072    .0095      .0104     .052       .1414     .1577      .1946
32   8,24 10,518,300 .0042     .0052      .0094     .058       .1631     .2024      .2133
Randomization tests
Full Randomization Test Problems (similar to t,F test)
    So ideal is to keep n1 = n2, but has practical limitations

What could be done to:
  N=32(8,24) : To bring back rejection level from 20% to 5% :
  Use BOOTSTRAPPING (Computationally intensive)
         Take scores at random (without replacement,let’s say 100 times) from larger groups
  to create          a sample of size equal to smaller group and do standard randomization test
         Each time noting whether H0 is rejected at 5% level.
         Increase is independent of differences in N
         Curves are averaged for different Variance ratios
         nominal level is controlled,
         ability to detect difference depends only on smaller n
         Resampling corrects too liberal behavior (test remains
         sensitive to true effects)

For F test, non-gaussian parent distributions: similar results

Caution: For equal    and unequal n: Resampling is
Conservative
Randomization tests
Full Randomization Test Problems : Bringing Computational cost under control
     Computations : (n1=10,n2=16, equal ) = C(26,10) = 26!/16!10! = 5,311,735 combinations
                    (larger in smaller cell) => resampling => 100 randomization tests each involves
                       C(20,10) = 184,756 combinations => Total 18,475,600 combinations

    Gill’s Algorithm : Gill(2007) used Fourier expansion to count extreme cases.
                              Under H0, all combinations of data in a randomization case are equally likely
                              Compute proportion of cases that is as or more extreme than observed data
                              one tail prob = P(T>t) + p(T=2) /2




                                                where tr is the value on rth combination




                                                where k = 2k’ –1, K’=1 to , and F(a) is imaginary part of a


Computational Cost brought down to practical level of a PC (little more costly than F,t but faster than full
   enumerations of all combinations
Conclusion
Assumptions of t, F tests create problems
Randomization test obviates that, but it has its own
problems
  Too conservative, Too liberal, and computationally
  intensive
  Liberal Bias can be removed by Bootstrapping, but it
  further makes it more computationally intensive
  Gill’s algorithm saves computational cost
  However algorithm is still asymmetric : No algorithm is
  known yet to remove Conservative bias
References

Fisher, Ronald A. “The Design of Experiments”. 8th ed. New
York: Hafner Publishing Company Inc., 1966.

Mewhort, D.J.K, Mathew Kelly and Johns Brendan
T.“Randomization tests and the unequal-N/unequal-variance
problem”

Gill, P. M.W. (2007). Efficient calculation of p-values in linear-
statistic permutation significance tests.Journal of Statistical
computation & Simulation, 77, 55-61.

Weitere Àhnliche Inhalte

Was ist angesagt?

Introduction to the t Statistic
Introduction to the t StatisticIntroduction to the t Statistic
Introduction to the t Statisticjasondroesch
 
Bayesian Neural Networks
Bayesian Neural NetworksBayesian Neural Networks
Bayesian Neural NetworksNatan Katz
 
To Explain, To Predict, or To Describe?
To Explain, To Predict, or To Describe?To Explain, To Predict, or To Describe?
To Explain, To Predict, or To Describe?Galit Shmueli
 
Hypothesis tests for one and two population variances ppt @ bec doms
Hypothesis tests for one and two population variances ppt @ bec domsHypothesis tests for one and two population variances ppt @ bec doms
Hypothesis tests for one and two population variances ppt @ bec domsBabasab Patil
 
Chapter 1 introduction to statistics for engineers 1 (1)
Chapter 1 introduction to statistics for engineers 1 (1)Chapter 1 introduction to statistics for engineers 1 (1)
Chapter 1 introduction to statistics for engineers 1 (1)abfisho
 
Spss vs excel
Spss vs excelSpss vs excel
Spss vs excelcalltutors
 
Development and evaluation of prediction models: pitfalls and solutions
Development and evaluation of prediction models: pitfalls and solutionsDevelopment and evaluation of prediction models: pitfalls and solutions
Development and evaluation of prediction models: pitfalls and solutionsMaarten van Smeden
 
Real world modified
Real world modifiedReal world modified
Real world modifiedStephen Senn
 
Chapter 6 part2-Introduction to Inference-Tests of Significance, Stating Hyp...
Chapter 6 part2-Introduction to Inference-Tests of Significance,  Stating Hyp...Chapter 6 part2-Introduction to Inference-Tests of Significance,  Stating Hyp...
Chapter 6 part2-Introduction to Inference-Tests of Significance, Stating Hyp...nszakir
 
Multivariate Analysis Techniques
Multivariate Analysis TechniquesMultivariate Analysis Techniques
Multivariate Analysis TechniquesMehul Gondaliya
 
The kolmogorov smirnov test
The kolmogorov smirnov testThe kolmogorov smirnov test
The kolmogorov smirnov testSubhradeep Mitra
 
Hypothesis testing
Hypothesis testingHypothesis testing
Hypothesis testingShameer P Hamsa
 
Fundamentals of Testing Hypothesis
Fundamentals of Testing HypothesisFundamentals of Testing Hypothesis
Fundamentals of Testing HypothesisYesica Adicondro
 
Measures of Central Tendency
Measures of Central TendencyMeasures of Central Tendency
Measures of Central Tendencyjasondroesch
 
Linear models for data science
Linear models for data scienceLinear models for data science
Linear models for data scienceBrad Klingenberg
 
What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...
What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...
What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...Pat Barlow
 
Lecture 5: Interval Estimation
Lecture 5: Interval Estimation Lecture 5: Interval Estimation
Lecture 5: Interval Estimation Marina Santini
 
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 testsEugene Yan Ziyou
 

Was ist angesagt? (20)

Introduction to the t Statistic
Introduction to the t StatisticIntroduction to the t Statistic
Introduction to the t Statistic
 
Bayesian Neural Networks
Bayesian Neural NetworksBayesian Neural Networks
Bayesian Neural Networks
 
To Explain, To Predict, or To Describe?
To Explain, To Predict, or To Describe?To Explain, To Predict, or To Describe?
To Explain, To Predict, or To Describe?
 
Hypothesis tests for one and two population variances ppt @ bec doms
Hypothesis tests for one and two population variances ppt @ bec domsHypothesis tests for one and two population variances ppt @ bec doms
Hypothesis tests for one and two population variances ppt @ bec doms
 
Chapter 1 introduction to statistics for engineers 1 (1)
Chapter 1 introduction to statistics for engineers 1 (1)Chapter 1 introduction to statistics for engineers 1 (1)
Chapter 1 introduction to statistics for engineers 1 (1)
 
Spss vs excel
Spss vs excelSpss vs excel
Spss vs excel
 
Development and evaluation of prediction models: pitfalls and solutions
Development and evaluation of prediction models: pitfalls and solutionsDevelopment and evaluation of prediction models: pitfalls and solutions
Development and evaluation of prediction models: pitfalls and solutions
 
Real world modified
Real world modifiedReal world modified
Real world modified
 
Chapter 6 part2-Introduction to Inference-Tests of Significance, Stating Hyp...
Chapter 6 part2-Introduction to Inference-Tests of Significance,  Stating Hyp...Chapter 6 part2-Introduction to Inference-Tests of Significance,  Stating Hyp...
Chapter 6 part2-Introduction to Inference-Tests of Significance, Stating Hyp...
 
Multivariate Analysis Techniques
Multivariate Analysis TechniquesMultivariate Analysis Techniques
Multivariate Analysis Techniques
 
The kolmogorov smirnov test
The kolmogorov smirnov testThe kolmogorov smirnov test
The kolmogorov smirnov test
 
Hypothesis testing
Hypothesis testingHypothesis testing
Hypothesis testing
 
Fundamentals of Testing Hypothesis
Fundamentals of Testing HypothesisFundamentals of Testing Hypothesis
Fundamentals of Testing Hypothesis
 
Measures of Central Tendency
Measures of Central TendencyMeasures of Central Tendency
Measures of Central Tendency
 
Linear models for data science
Linear models for data scienceLinear models for data science
Linear models for data science
 
What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...
What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...
What's Significant? Hypothesis Testing, Effect Size, Confidence Intervals, & ...
 
Lecture 5: Interval Estimation
Lecture 5: Interval Estimation Lecture 5: Interval Estimation
Lecture 5: Interval Estimation
 
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
 
Basics of Hypothesis Testing
Basics of Hypothesis Testing  Basics of Hypothesis Testing
Basics of Hypothesis Testing
 
Chap010
Chap010Chap010
Chap010
 

Andere mochten auch

Andere mochten auch (6)

Randomization
Randomization Randomization
Randomization
 
Randomisation techniques
Randomisation techniquesRandomisation techniques
Randomisation techniques
 
Plant and animal viruse
Plant and animal virusePlant and animal viruse
Plant and animal viruse
 
Ch 6 randomization
Ch 6 randomizationCh 6 randomization
Ch 6 randomization
 
Pharmacogenomics
PharmacogenomicsPharmacogenomics
Pharmacogenomics
 
Pharmacogenomics
PharmacogenomicsPharmacogenomics
Pharmacogenomics
 

Ähnlich wie Randomization Tests

Approximate ANCOVA
Approximate ANCOVAApproximate ANCOVA
Approximate ANCOVAStephen Senn
 
Statistics practice for finalBe sure to review the following.docx
Statistics practice for finalBe sure to review the following.docxStatistics practice for finalBe sure to review the following.docx
Statistics practice for finalBe sure to review the following.docxdessiechisomjj4
 
jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj
jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj
jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjMdSazolAhmmed
 
Hypothesis testing
Hypothesis testingHypothesis testing
Hypothesis testingDenni Domingo
 
Non parametrics tests
Non parametrics testsNon parametrics tests
Non parametrics testsrodrick koome
 
Solution to the practice test ch 8 hypothesis testing ch 9 two populations
Solution to the practice test ch 8 hypothesis testing ch 9 two populationsSolution to the practice test ch 8 hypothesis testing ch 9 two populations
Solution to the practice test ch 8 hypothesis testing ch 9 two populationsLong Beach City College
 
Multiple estimators for Monte Carlo approximations
Multiple estimators for Monte Carlo approximationsMultiple estimators for Monte Carlo approximations
Multiple estimators for Monte Carlo approximationsChristian Robert
 
Part 1 of 16 - Question 1 of 231.0 PointsThe data pres.docx
Part 1 of 16 - Question 1 of 231.0 PointsThe data pres.docxPart 1 of 16 - Question 1 of 231.0 PointsThe data pres.docx
Part 1 of 16 - Question 1 of 231.0 PointsThe data pres.docxherbertwilson5999
 
CFA Fit Statistics
CFA Fit StatisticsCFA Fit Statistics
CFA Fit Statisticsnicolalritter
 
statistics-for-analytical-chemistry (1).ppt
statistics-for-analytical-chemistry (1).pptstatistics-for-analytical-chemistry (1).ppt
statistics-for-analytical-chemistry (1).pptHalilIbrahimUlusoy
 
probability ch 6 ppt_1_1.pptx
probability ch 6 ppt_1_1.pptxprobability ch 6 ppt_1_1.pptx
probability ch 6 ppt_1_1.pptxYeMinThant4
 
Lecture 07 Category Shaoqi Rao Rev
Lecture 07 Category Shaoqi Rao RevLecture 07 Category Shaoqi Rao Rev
Lecture 07 Category Shaoqi Rao RevSumit Prajapati
 
Two Variances or Standard Deviations
Two Variances or Standard DeviationsTwo Variances or Standard Deviations
Two Variances or Standard DeviationsLong Beach City College
 

Ähnlich wie Randomization Tests (20)

Approximate ANCOVA
Approximate ANCOVAApproximate ANCOVA
Approximate ANCOVA
 
Statistics practice for finalBe sure to review the following.docx
Statistics practice for finalBe sure to review the following.docxStatistics practice for finalBe sure to review the following.docx
Statistics practice for finalBe sure to review the following.docx
 
Chapter12
Chapter12Chapter12
Chapter12
 
Two Proportions
Two Proportions  Two Proportions
Two Proportions
 
jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj
jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj
jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj
 
STATISTIC ESTIMATION
STATISTIC ESTIMATIONSTATISTIC ESTIMATION
STATISTIC ESTIMATION
 
Hypothesis testing
Hypothesis testingHypothesis testing
Hypothesis testing
 
Non parametrics tests
Non parametrics testsNon parametrics tests
Non parametrics tests
 
9618821.pdf
9618821.pdf9618821.pdf
9618821.pdf
 
9618821.ppt
9618821.ppt9618821.ppt
9618821.ppt
 
Solution to the practice test ch 8 hypothesis testing ch 9 two populations
Solution to the practice test ch 8 hypothesis testing ch 9 two populationsSolution to the practice test ch 8 hypothesis testing ch 9 two populations
Solution to the practice test ch 8 hypothesis testing ch 9 two populations
 
Mech ma6452 snm_notes
Mech ma6452 snm_notesMech ma6452 snm_notes
Mech ma6452 snm_notes
 
Multiple estimators for Monte Carlo approximations
Multiple estimators for Monte Carlo approximationsMultiple estimators for Monte Carlo approximations
Multiple estimators for Monte Carlo approximations
 
Part 1 of 16 - Question 1 of 231.0 PointsThe data pres.docx
Part 1 of 16 - Question 1 of 231.0 PointsThe data pres.docxPart 1 of 16 - Question 1 of 231.0 PointsThe data pres.docx
Part 1 of 16 - Question 1 of 231.0 PointsThe data pres.docx
 
CFA Fit Statistics
CFA Fit StatisticsCFA Fit Statistics
CFA Fit Statistics
 
statistics-for-analytical-chemistry (1).ppt
statistics-for-analytical-chemistry (1).pptstatistics-for-analytical-chemistry (1).ppt
statistics-for-analytical-chemistry (1).ppt
 
probability ch 6 ppt_1_1.pptx
probability ch 6 ppt_1_1.pptxprobability ch 6 ppt_1_1.pptx
probability ch 6 ppt_1_1.pptx
 
Lecture 07 Category Shaoqi Rao Rev
Lecture 07 Category Shaoqi Rao RevLecture 07 Category Shaoqi Rao Rev
Lecture 07 Category Shaoqi Rao Rev
 
Two Variances or Standard Deviations
Two Variances or Standard DeviationsTwo Variances or Standard Deviations
Two Variances or Standard Deviations
 
SNM-1.pdf
SNM-1.pdfSNM-1.pdf
SNM-1.pdf
 

Mehr von Ajay Dhamija

fm3-05-301 (copy).pdf
fm3-05-301 (copy).pdffm3-05-301 (copy).pdf
fm3-05-301 (copy).pdfAjay Dhamija
 
Carbon Finance
Carbon FinanceCarbon Finance
Carbon FinanceAjay Dhamija
 
Ethical hacking & Information Security
Ethical hacking & Information SecurityEthical hacking & Information Security
Ethical hacking & Information SecurityAjay Dhamija
 
Karmarkar's Algorithm For Linear Programming Problem
Karmarkar's Algorithm For Linear Programming ProblemKarmarkar's Algorithm For Linear Programming Problem
Karmarkar's Algorithm For Linear Programming ProblemAjay Dhamija
 
Verizon - A Case Study
Verizon - A Case StudyVerizon - A Case Study
Verizon - A Case StudyAjay Dhamija
 
Dabur India Ltd - A Case Study
Dabur India Ltd  - A Case StudyDabur India Ltd  - A Case Study
Dabur India Ltd - A Case StudyAjay Dhamija
 
Non Banking Financial Company
Non Banking Financial CompanyNon Banking Financial Company
Non Banking Financial CompanyAjay Dhamija
 
The Financial Sector Reforms in India
The Financial Sector Reforms in IndiaThe Financial Sector Reforms in India
The Financial Sector Reforms in IndiaAjay Dhamija
 
Hosting Inviting Introduction Guest Relations
Hosting Inviting Introduction Guest RelationsHosting Inviting Introduction Guest Relations
Hosting Inviting Introduction Guest RelationsAjay Dhamija
 
Global Fiancial Meltdown of 2007
Global Fiancial Meltdown of 2007Global Fiancial Meltdown of 2007
Global Fiancial Meltdown of 2007Ajay Dhamija
 
IRT - Item response Theory
IRT - Item response TheoryIRT - Item response Theory
IRT - Item response TheoryAjay Dhamija
 
Goody Research - Research Methods Flaws
Goody Research - Research Methods FlawsGoody Research - Research Methods Flaws
Goody Research - Research Methods FlawsAjay Dhamija
 
Power Analysis and Sample Size Determination
Power Analysis and Sample Size DeterminationPower Analysis and Sample Size Determination
Power Analysis and Sample Size DeterminationAjay Dhamija
 
Research Design
Research DesignResearch Design
Research DesignAjay Dhamija
 

Mehr von Ajay Dhamija (15)

fm3-05-301 (copy).pdf
fm3-05-301 (copy).pdffm3-05-301 (copy).pdf
fm3-05-301 (copy).pdf
 
Carbon Finance
Carbon FinanceCarbon Finance
Carbon Finance
 
Ethical hacking & Information Security
Ethical hacking & Information SecurityEthical hacking & Information Security
Ethical hacking & Information Security
 
Karmarkar's Algorithm For Linear Programming Problem
Karmarkar's Algorithm For Linear Programming ProblemKarmarkar's Algorithm For Linear Programming Problem
Karmarkar's Algorithm For Linear Programming Problem
 
Verizon - A Case Study
Verizon - A Case StudyVerizon - A Case Study
Verizon - A Case Study
 
Dabur India Ltd - A Case Study
Dabur India Ltd  - A Case StudyDabur India Ltd  - A Case Study
Dabur India Ltd - A Case Study
 
Non Banking Financial Company
Non Banking Financial CompanyNon Banking Financial Company
Non Banking Financial Company
 
The Financial Sector Reforms in India
The Financial Sector Reforms in IndiaThe Financial Sector Reforms in India
The Financial Sector Reforms in India
 
Hosting Inviting Introduction Guest Relations
Hosting Inviting Introduction Guest RelationsHosting Inviting Introduction Guest Relations
Hosting Inviting Introduction Guest Relations
 
TRIZ
TRIZ TRIZ
TRIZ
 
Global Fiancial Meltdown of 2007
Global Fiancial Meltdown of 2007Global Fiancial Meltdown of 2007
Global Fiancial Meltdown of 2007
 
IRT - Item response Theory
IRT - Item response TheoryIRT - Item response Theory
IRT - Item response Theory
 
Goody Research - Research Methods Flaws
Goody Research - Research Methods FlawsGoody Research - Research Methods Flaws
Goody Research - Research Methods Flaws
 
Power Analysis and Sample Size Determination
Power Analysis and Sample Size DeterminationPower Analysis and Sample Size Determination
Power Analysis and Sample Size Determination
 
Research Design
Research DesignResearch Design
Research Design
 

KĂŒrzlich hochgeladen

Kenya’s Coconut Value Chain by Gatsby Africa
Kenya’s Coconut Value Chain by Gatsby AfricaKenya’s Coconut Value Chain by Gatsby Africa
Kenya’s Coconut Value Chain by Gatsby Africaictsugar
 
Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Seta Wicaksana
 
Call US-88OO1O2216 Call Girls In Mahipalpur Female Escort Service
Call US-88OO1O2216 Call Girls In Mahipalpur Female Escort ServiceCall US-88OO1O2216 Call Girls In Mahipalpur Female Escort Service
Call US-88OO1O2216 Call Girls In Mahipalpur Female Escort Servicecallgirls2057
 
8447779800, Low rate Call girls in Uttam Nagar Delhi NCR
8447779800, Low rate Call girls in Uttam Nagar Delhi NCR8447779800, Low rate Call girls in Uttam Nagar Delhi NCR
8447779800, Low rate Call girls in Uttam Nagar Delhi NCRashishs7044
 
Investment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy CheruiyotInvestment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy Cheruiyotictsugar
 
Cybersecurity Awareness Training Presentation v2024.03
Cybersecurity Awareness Training Presentation v2024.03Cybersecurity Awareness Training Presentation v2024.03
Cybersecurity Awareness Training Presentation v2024.03DallasHaselhorst
 
Kenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith PereraKenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith Pereraictsugar
 
Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Peter Ward
 
Market Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 EditionMarket Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 EditionMintel Group
 
8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCR8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCRashishs7044
 
Digital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdfDigital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdfJos Voskuil
 
(Best) ENJOY Call Girls in Faridabad Ex | 8377087607
(Best) ENJOY Call Girls in Faridabad Ex | 8377087607(Best) ENJOY Call Girls in Faridabad Ex | 8377087607
(Best) ENJOY Call Girls in Faridabad Ex | 8377087607dollysharma2066
 
Organizational Structure Running A Successful Business
Organizational Structure Running A Successful BusinessOrganizational Structure Running A Successful Business
Organizational Structure Running A Successful BusinessSeta Wicaksana
 
Call Us đŸ“Č8800102216📞 Call Girls In DLF City Gurgaon
Call Us đŸ“Č8800102216📞 Call Girls In DLF City GurgaonCall Us đŸ“Č8800102216📞 Call Girls In DLF City Gurgaon
Call Us đŸ“Č8800102216📞 Call Girls In DLF City Gurgaoncallgirls2057
 
Independent Call Girls Andheri Nightlaila 9967584737
Independent Call Girls Andheri Nightlaila 9967584737Independent Call Girls Andheri Nightlaila 9967584737
Independent Call Girls Andheri Nightlaila 9967584737Riya Pathan
 
8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCR8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCRashishs7044
 
International Business Environments and Operations 16th Global Edition test b...
International Business Environments and Operations 16th Global Edition test b...International Business Environments and Operations 16th Global Edition test b...
International Business Environments and Operations 16th Global Edition test b...ssuserf63bd7
 

KĂŒrzlich hochgeladen (20)

Kenya’s Coconut Value Chain by Gatsby Africa
Kenya’s Coconut Value Chain by Gatsby AfricaKenya’s Coconut Value Chain by Gatsby Africa
Kenya’s Coconut Value Chain by Gatsby Africa
 
Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...
 
Call US-88OO1O2216 Call Girls In Mahipalpur Female Escort Service
Call US-88OO1O2216 Call Girls In Mahipalpur Female Escort ServiceCall US-88OO1O2216 Call Girls In Mahipalpur Female Escort Service
Call US-88OO1O2216 Call Girls In Mahipalpur Female Escort Service
 
Corporate Profile 47Billion Information Technology
Corporate Profile 47Billion Information TechnologyCorporate Profile 47Billion Information Technology
Corporate Profile 47Billion Information Technology
 
8447779800, Low rate Call girls in Uttam Nagar Delhi NCR
8447779800, Low rate Call girls in Uttam Nagar Delhi NCR8447779800, Low rate Call girls in Uttam Nagar Delhi NCR
8447779800, Low rate Call girls in Uttam Nagar Delhi NCR
 
Investment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy CheruiyotInvestment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy Cheruiyot
 
Cybersecurity Awareness Training Presentation v2024.03
Cybersecurity Awareness Training Presentation v2024.03Cybersecurity Awareness Training Presentation v2024.03
Cybersecurity Awareness Training Presentation v2024.03
 
Kenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith PereraKenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith Perera
 
Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...
 
Market Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 EditionMarket Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 Edition
 
8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCR8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCR
 
Digital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdfDigital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdf
 
(Best) ENJOY Call Girls in Faridabad Ex | 8377087607
(Best) ENJOY Call Girls in Faridabad Ex | 8377087607(Best) ENJOY Call Girls in Faridabad Ex | 8377087607
(Best) ENJOY Call Girls in Faridabad Ex | 8377087607
 
Organizational Structure Running A Successful Business
Organizational Structure Running A Successful BusinessOrganizational Structure Running A Successful Business
Organizational Structure Running A Successful Business
 
Call Us đŸ“Č8800102216📞 Call Girls In DLF City Gurgaon
Call Us đŸ“Č8800102216📞 Call Girls In DLF City GurgaonCall Us đŸ“Č8800102216📞 Call Girls In DLF City Gurgaon
Call Us đŸ“Č8800102216📞 Call Girls In DLF City Gurgaon
 
Independent Call Girls Andheri Nightlaila 9967584737
Independent Call Girls Andheri Nightlaila 9967584737Independent Call Girls Andheri Nightlaila 9967584737
Independent Call Girls Andheri Nightlaila 9967584737
 
Call Us ➄9319373153▻Call Girls In North Goa
Call Us ➄9319373153▻Call Girls In North GoaCall Us ➄9319373153▻Call Girls In North Goa
Call Us ➄9319373153▻Call Girls In North Goa
 
8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCR8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCR
 
International Business Environments and Operations 16th Global Edition test b...
International Business Environments and Operations 16th Global Edition test b...International Business Environments and Operations 16th Global Edition test b...
International Business Environments and Operations 16th Global Edition test b...
 
No-1 Call Girls In Goa 93193 VIP 73153 Escort service In North Goa Panaji, Ca...
No-1 Call Girls In Goa 93193 VIP 73153 Escort service In North Goa Panaji, Ca...No-1 Call Girls In Goa 93193 VIP 73153 Escort service In North Goa Panaji, Ca...
No-1 Call Girls In Goa 93193 VIP 73153 Escort service In North Goa Panaji, Ca...
 

Randomization Tests

  • 1. Randomization Tests – unequal-N, unequal-σ problem AK Dhamija
  • 2. Agenda Assumptions of t, F tests Randomization tests Problems of Randomization Test Too liberal Too conservative Computationally Intensive Solving the problems Resampling Gill’s algorithm
  • 3. Assumptions of t, F tests The two samples are each drawn from normal distributions. The two samples are drawn randomly from their respective populations. RANDOMIZATION TESTS TACKLE THESE UNREALISTIC ASSUMPTIONS
  • 4. Randomization tests An Example Comparing t-Test and Randomization Test Results Two fertilizers (A and B) that are randomly applied to a type of sunflower seed. The maximum heights reached (in feet) are recorded after some time period. All Other Factors are constant Null hypothesis : no difference between fertilizers A and B with respect to sunflower height. Alternative hypothesis : fertilizer A is superior to fertilizer B on average with respect to sunflower height. Sample Fertilizer Height (ft) 1 A 9.9 2 B 9.6 3 B 9.7 4 B 9.4 5 A 10.1 6 B 9.5 7 A 9.9 8 B 9.6 Total 462 (11 !/ 5! 6!) permutations 9 A 9.5 5 of the 462 showed mean difference of 9.920 – 9.533 = 0.387 10 A 10.2 p-value = 5/ 462 = 0.0108 => Reject H0 (t-test also rejects) 11 B 9.4 = > fertilizer A outperforms fertilizer B So t-test provides reasonably good approximation to randomization test
  • 5. Randomization tests Randomization Tests do not consider normality, random sampling, equal variances, or other assumptions. The conclusion was based solely on the observed results, and the fact that the fertilizers were randomly assigned. Why randomization tests then are not widely used, nor addressed in many statistical texts. The number of computations with larger sample sizes becomes astronomical With two samples, each of size 30, there are over 1.18 * 1017 possible permutations! But randomization tests becomes sensitive to heteroscedasticity when the cells are unequal in size Approximate randomization Tests (selecting few combinations) Unstable – (statistics may vary) Unreplicable
  • 6. Randomization tests Full Randomization Test Problems (similar to t,F test) Too conservative if larger cells have larger variances (large effect is required for significance) Too liberal if smaller cells have larger variances (exaggerates the true difference) Variance Ratios N n1,n2 C(N,n1) 1:10 1:4 1:2 1:1 2:1 4:1 10:1 16 8,8 12,870 .0744 .0585 .0594 .045 .0616 .0464 .0656 20 8,14 125,970 .0312 .03 .0319 .058 .0921 .0984 .1152 24 8,16 735,471 .0156 .0158 .0181 .0468 .1222 .1304 .1618 28 8,20 3,108,105 .0072 .0095 .0104 .052 .1414 .1577 .1946 32 8,24 10,518,300 .0042 .0052 .0094 .058 .1631 .2024 .2133
  • 7. Randomization tests Full Randomization Test Problems (similar to t,F test) So ideal is to keep n1 = n2, but has practical limitations What could be done to: N=32(8,24) : To bring back rejection level from 20% to 5% : Use BOOTSTRAPPING (Computationally intensive) Take scores at random (without replacement,let’s say 100 times) from larger groups to create a sample of size equal to smaller group and do standard randomization test Each time noting whether H0 is rejected at 5% level. Increase is independent of differences in N Curves are averaged for different Variance ratios nominal level is controlled, ability to detect difference depends only on smaller n Resampling corrects too liberal behavior (test remains sensitive to true effects) For F test, non-gaussian parent distributions: similar results Caution: For equal and unequal n: Resampling is Conservative
  • 8. Randomization tests Full Randomization Test Problems : Bringing Computational cost under control Computations : (n1=10,n2=16, equal ) = C(26,10) = 26!/16!10! = 5,311,735 combinations (larger in smaller cell) => resampling => 100 randomization tests each involves C(20,10) = 184,756 combinations => Total 18,475,600 combinations Gill’s Algorithm : Gill(2007) used Fourier expansion to count extreme cases. Under H0, all combinations of data in a randomization case are equally likely Compute proportion of cases that is as or more extreme than observed data one tail prob = P(T>t) + p(T=2) /2 where tr is the value on rth combination where k = 2k’ –1, K’=1 to , and F(a) is imaginary part of a Computational Cost brought down to practical level of a PC (little more costly than F,t but faster than full enumerations of all combinations
  • 9. Conclusion Assumptions of t, F tests create problems Randomization test obviates that, but it has its own problems Too conservative, Too liberal, and computationally intensive Liberal Bias can be removed by Bootstrapping, but it further makes it more computationally intensive Gill’s algorithm saves computational cost However algorithm is still asymmetric : No algorithm is known yet to remove Conservative bias
  • 10. References Fisher, Ronald A. “The Design of Experiments”. 8th ed. New York: Hafner Publishing Company Inc., 1966. Mewhort, D.J.K, Mathew Kelly and Johns Brendan T.“Randomization tests and the unequal-N/unequal-variance problem” Gill, P. M.W. (2007). Efficient calculation of p-values in linear- statistic permutation significance tests.Journal of Statistical computation & Simulation, 77, 55-61.