Hypothesis testing. Null hypothesis H 0 and alternative hypothesis H 1.

Save this PDF as:
 WORD  PNG  TXT  JPG

Size: px
Start display at page:

Download "Hypothesis testing. Null hypothesis H 0 and alternative hypothesis H 1."

Transcription

1 Hypothesis testing Null hypothesis H 0 and alternative hypothesis H 1.

2 Hypothesis testing Null hypothesis H 0 and alternative hypothesis H 1. Simple and compound hypotheses. Simple : the probabilistic model is specified completely. Compound : the probabilistic model is not specified completely (generally it will contain parameters to be estimated).

3 Hypothesis testing Null hypothesis H 0 and alternative hypothesis H 1. Simple and compound hypotheses. Simple : the probabilistic model is specified completely. Compound : the probabilistic model is not specified completely (generally it will contain parameters to be estimated). Example 1: we want to test whether data are compatible with the assumption that their true mean is µ 0. Then it could be set as: H 0 : X 1,..., X n N(µ 0, σ0 2 ) and independent. [simple if σ 2 known]; H 1 : X 1,..., X n N(µ, σ 2 0 ) and independent, with µ µ 0. [compound]

4 Rejection region Example 2: we have two groups, and we wish to test whether they can be considered as samples from the same population, or from two populations with different means. General assumption: X 1,..., X n N(µ X, σ 2 ), Y 1,..., Y m N(µ Y, σ 2 ), and independent. H 0 : µ X = µ Y, σ 2 > 0. H 1 : µ X µ Y, σ 2 > 0. Both are compound, but H 0 is simpler than H 1. How does a test work? We select a rejection region C: if data fall in C, we reject H 0 (and accept H 1 ); if data do not fall in C, we accept (do not reject) H 0.

5 Errors of first and second species Error of first species: rejecting H 0 if H 0 is true; Error of second species: accepting H 0 if H 1 is true. A smaller rejection region C decreases error of first species, but increases those of second species; a larger C vice versa.

6 Errors of first and second species Error of first species: rejecting H 0 if H 0 is true; Error of second species: accepting H 0 if H 1 is true. A smaller rejection region C decreases error of first species, but increases those of second species; a larger C vice versa. Solution? We choose a level (the risk we take of errors of 1st species) α [often 5%] and select a region s.t. P(X C H 0 ) = α (probability of error of 1st species).

7 Errors of first and second species Error of first species: rejecting H 0 if H 0 is true; Error of second species: accepting H 0 if H 1 is true. A smaller rejection region C decreases error of first species, but increases those of second species; a larger C vice versa. Solution? We choose a level (the risk we take of errors of 1st species) α [often 5%] and select a region s.t. P(X C H 0 ) = α (probability of error of 1st species). A test will have a certain power (the probability of rejecting H 0 when indeed H 1 is true).

8 Errors of first and second species Error of first species: rejecting H 0 if H 0 is true; Error of second species: accepting H 0 if H 1 is true. A smaller rejection region C decreases error of first species, but increases those of second species; a larger C vice versa. Solution? We choose a level (the risk we take of errors of 1st species) α [often 5%] and select a region s.t. P(X C H 0 ) = α (probability of error of 1st species). A test will have a certain power (the probability of rejecting H 0 when indeed H 1 is true). Generally, we are not able to compute exactly the power of a test (it depends on exact parameter value), but theory selects the most powerful test within a certain level.

9 One-sample test on the mean Example 1: (with σ 2 unknown). H 0 : X 1,..., X n N(µ 0, σ 2 ) and independent, where σ 2 > 0. H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ µ 0, σ 2 > 0.

10 One-sample test on the mean Example 1: (with σ 2 unknown). H 0 : X 1,..., X n N(µ 0, σ 2 ) and independent, where σ 2 > 0. H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ µ 0, σ 2 > 0. The test quantity used is T = X µ 0 S/ n where S 2 is the sample variance. It is natural (and can be justified rigorously) to reject H 0 when T is far away from 0.

11 One-sample test on the mean Example 1: (with σ 2 unknown). H 0 : X 1,..., X n N(µ 0, σ 2 ) and independent, where σ 2 > 0. H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ µ 0, σ 2 > 0. The test quantity used is T = X µ 0 S/ n where S 2 is the sample variance. It is natural (and can be justified rigorously) to reject H 0 when T is far away from 0. How far is far?

12 One-sample test on the mean Example 1: (with σ 2 unknown). H 0 : X 1,..., X n N(µ 0, σ 2 ) and independent, where σ 2 > 0. H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ µ 0, σ 2 > 0. The test quantity used is T = X µ 0 S/ n where S 2 is the sample variance. It is natural (and can be justified rigorously) to reject H 0 when T is far away from 0. How far is far? Under H 0, T follows a t(n 1) distribution. Then we find t α s.t. P( t(n 1) > t α ) = α. Reject H 0 if T > t α, accept it otherwise.

13 Computations It is expected that mean body temperature is 37.0 C. Temperatures were obtained rom 25 individuals: Are these compatible with µ = 37.0?

14 Computations It is expected that mean body temperature is 37.0 C. Temperatures were obtained rom 25 individuals: Are these compatible with µ = 37.0? X = , S = 0.374, T =

15 Computations It is expected that mean body temperature is 37.0 C. Temperatures were obtained rom 25 individuals: Are these compatible with µ = 37.0? p-value = X = , S = 0.374, T =

16 Computations It is expected that mean body temperature is 37.0 C. Temperatures were obtained rom 25 individuals: Are these compatible with µ = 37.0? X = , S = 0.374, T = p-value = (this means that there is about 56% chance of getting a higher (in absolute value) value of T if H 0 is true). [Many statistical programs (e.g. R) return the p-value; the user will know how it compares with α she/he has chosen] Here we accept µ = 37.0 at any reasonable level of α.

17 One-sample test on the mean. II If [unilateral alternative] H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ > µ 0, σ 2 > 0, then the rejection region is for T positive and large. Hence we find t α s.t. P(t(n 1) > t α) = α. Reject H 0 if T > t α, accept it otherwise. Vice versa if the alternative hypothesis is µ < µ 0.

18 One-sample test on the mean. II If [unilateral alternative] H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ > µ 0, σ 2 > 0, then the rejection region is for T positive and large. Hence we find t α s.t. P(t(n 1) > t α) = α. Reject H 0 if T > t α, accept it otherwise. Vice versa if the alternative hypothesis is µ < µ 0. In the example, if H 1 had been µ < 37, p-value = P(t(24) < ) = 0.281

19 One-sample test on the mean. II If [unilateral alternative] H 1 : X 1,..., X n N(µ, σ 2 ) and independent, where µ > µ 0, σ 2 > 0, then the rejection region is for T positive and large. Hence we find t α s.t. P(t(n 1) > t α) = α. Reject H 0 if T > t α, accept it otherwise. Vice versa if the alternative hypothesis is µ < µ 0. In the example, if H 1 had been µ < 37, p-value = P(t(24) < ) = Observations: unilateral alternatives make it easier rejecting the null hypothesis (hence they are seldom used). In practice, border-line results suggest further research.

20 Test on equality of the means Independent samples (e.g. 2 groups with different treatments) X 1,..., X n N(µ X, σ 2 ), Y 1,..., Y m N(µ Y, σ 2 ), and independent. H 0 : µ X = µ Y, σ 2 > 0. H 1 : µ X µ Y, σ 2 > 0.

21 Test on equality of the means Independent samples (e.g. 2 groups with different treatments) X 1,..., X n N(µ X, σ 2 ), Y 1,..., Y m N(µ Y, σ 2 ), and independent. H 0 : µ X = µ Y, σ 2 > 0. H 1 : µ X µ Y, σ 2 > 0. Under H 0 Y i X j N(0, 2σ 2 ) for all i = 1... m, j = 1... n. Hence we test whether the true mean of all these variables is 0.

22 Test on equality of the means Independent samples (e.g. 2 groups with different treatments) X 1,..., X n N(µ X, σ 2 ), Y 1,..., Y m N(µ Y, σ 2 ), and independent. H 0 : µ X = µ Y, σ 2 > 0. H 1 : µ X µ Y, σ 2 > 0. Under H 0 Y i X j N(0, 2σ 2 ) for all i = 1... m, j = 1... n. Hence we test whether the true mean of all these variables is 0. Estimate of σ 2 : SX 2,Y = 1 ( (n 1)S 2 n + m 2 X + (m 1)SY 2 ) weighted mean of S 2 X (variance estimated on X i) and S 2 Y.

23 Test on equality of the means Independent samples (e.g. 2 groups with different treatments) X 1,..., X n N(µ X, σ 2 ), Y 1,..., Y m N(µ Y, σ 2 ), and independent. H 0 : µ X = µ Y, σ 2 > 0. H 1 : µ X µ Y, σ 2 > 0. Under H 0 Y i X j N(0, 2σ 2 ) for all i = 1... m, j = 1... n. Hence we test whether the true mean of all these variables is 0. Estimate of σ 2 : SX 2,Y = 1 ( (n 1)S 2 n + m 2 X + (m 1)SY 2 ) weighted mean of SX 2 (variance estimated on X i) and SY 2. Ȳ X Finally, t-test on T =. s X,Y 1 n + 1 m

24 Test on equality of the means Independent samples (e.g. 2 groups with different treatments) X 1,..., X n N(µ X, σ 2 ), Y 1,..., Y m N(µ Y, σ 2 ), and independent. H 0 : µ X = µ Y, σ 2 > 0. H 1 : µ X µ Y, σ 2 > 0. Under H 0 Y i X j N(0, 2σ 2 ) for all i = 1... m, j = 1... n. Hence we test whether the true mean of all these variables is 0. Estimate of σ 2 : SX 2,Y = 1 ( (n 1)S 2 n + m 2 X + (m 1)SY 2 ) weighted mean of SX 2 (variance estimated on X i) and SY 2. Ȳ X Finally, t-test on T =. s X,Y 1 n + 1 m Assumptions: normality, independence, equality of variances (should be checked [sometimes variable transformations help]).

25 Test of equality of means with independent samples: computations Annual flow of the river Nile at Ashwan has been measured We ask whether the mean of the former 50 years is the same as the mean of the latter 50 years.

26 Test of equality of means with independent samples: computations Annual flow of the river Nile at Ashwan has been measured We ask whether the mean of the former 50 years is the same as the mean of the latter 50 years. Treat each year as an independent sample, though assumption questionable

27 Test of equality of means with independent samples: computations Annual flow of the river Nile at Ashwan has been measured We ask whether the mean of the former 50 years is the same as the mean of the latter 50 years. X (mean of years 1 50) = , Ȳ (mean of years ) = SX 2 = 37, 140, S Y 2 = 12, 106. S X 2,Y = 24, 623.

28 Test of equality of means with independent samples: computations Annual flow of the river Nile at Ashwan has been measured We ask whether the mean of the former 50 years is the same as the mean of the latter 50 years. X (mean of years 1 50) = , Ȳ (mean of years ) = SX 2 = 37, 140, S Y 2 = 12, 106. S X 2,Y = 24, 623. Hence T = , p-value = Reject µ X = µ Y.

29 Test of equality of means with independent samples: computations Annual flow of the river Nile at Ashwan has been measured We ask whether the mean of the former 50 years is the same as the mean of the latter 50 years. X (mean of years 1 50) = , Ȳ (mean of years ) = S 2 X = 37, 140, S 2 Y = 12, 106. S 2 X,Y = 24, 623. Hence T = , p-value = Reject µ X = µ Y. One may question equality of variances... A correction of the t-test works without equal variances: t = , df = , p value =

30 Test on equality of the means. II Paired samples (e.g. same individuals before/after treatment) General assumption Y i X i N(µ, σ 2 ), i = 1... n. No assumption on X i and Y i, but only on their differences (the effect of treatment). H 0 : µ = 0, σ 2 > 0. H 1 : µ 0, σ 2 > 0. This is simply a test that the true mean of W = Y X is 0. It will be easier rejecting H 0 because generally s Y X is much smaller than 2s X,Y. Basic assumption: the effect of treatment is additive (does not depend on the original value of X i ).

31 Test on binomial Observations: n trials with k successes. Are these compatible with probability of success p 0? H 0 : X Bin(n, p 0 ) H 1 : X Bin(n, p) with p p 0. Simplest method: under H 0 by central limit theorem Z = X np 0 np0 (1 p 0 ) N(0, 1) for n large [rule np 0, n(1 p 0 ) > 5]. Set α find z α s.t. P( N(0, 1) > z α ) = α. If Z > z α, reject H 0 ; otherwise accept it. Exact method: make computations with binomial distribution. Note. All previous cases could be studied through confidence intervals. Some authors prefer using only confidence intervals.

32 Chi-square test General chi-square test: We have k types of events that can occur in each trial, with a priori probabilities for them p 0 1,..., p 0 k with p p 0 k = 1. After n trials, we observe n 1 events of type 1, n 2 of 2,..., n k of k, with n n k = n. Are data compatible with expectations? (k = 2 is binomial) Classical test: chi-square: Set E i = np 0 i (expected number of events of type i under H 0 ), X 2 = k (n i E i ) 2 i=1 E i χ 2 (k 1) for n large. Find c α s.t. P(χ 2 (k 1) > c α ) = α. If X 2 > c α, reject H 0 ; accept it otherwise.

33 Chi-square for data fit to a distribution The values p1 0,..., p0 k can be those arising from some distribution. Often the distribution will contain parameters to be estimated (e.g. λ of Poisson). One can use the chi-square: if m parameters are estimated, X 2 χ 2 (k m 1) (of course m < k 1). Example: data (Von Bartkiewicz, 1898) on Prussians soldiers kicked to death by horses: i (deaths) n i (number of corps/years) Total 200

34 Chi-square example (continued) Estimate λ with the sample mean ˆλ = ( )/200 = Compute E i. Join the classes 3 (rule of thumb: E i 5) to obtain: i n i Ê i Total 200 Compute X lower than relevant quantiles of χ 2 (2).

35 Chi-square test of independence A classical use of chi-square is when we observe two qualitative variables. H 0 : variables are independent; H 1 : they are not independent. k levels of X and l levels of Y (if k = l = 2, a 2 2 contingency table). Data are n ij (# of observations with X = i (i = 1... k) and Y = j (j = 1... l). H 0 : P(X = i, Y = j) = p i q j for all i and j p i, i = 1... k 1, q j, j = 1... l 1 to be estimated from data. H 1 : P(X = i, Y = j) p i q j. Examples and computations to be briefly presented in next lecture.

Wording of Final Conclusion. Slide 1

Wording of Final Conclusion. Slide 1 Wording of Final Conclusion Slide 1 8.3: Assumptions for Testing Slide 2 Claims About Population Means 1) The sample is a simple random sample. 2) The value of the population standard deviation σ is known

More information

Hence, multiplying by 12, the 95% interval for the hourly rate is (965, 1435)

Hence, multiplying by 12, the 95% interval for the hourly rate is (965, 1435) Confidence Intervals for Poisson data For an observation from a Poisson distribution, we have σ 2 = λ. If we observe r events, then our estimate ˆλ = r : N(λ, λ) If r is bigger than 20, we can use this

More information

Lecture 2: Statistical Estimation and Testing

Lecture 2: Statistical Estimation and Testing Bioinformatics: In-depth PROBABILITY AND STATISTICS Spring Semester 2012 Lecture 2: Statistical Estimation and Testing Stefanie Muff (stefanie.muff@ifspm.uzh.ch) 1 Problems in statistics 2 The three main

More information

Lecture 1: t tests and CLT

Lecture 1: t tests and CLT Lecture 1: t tests and CLT http://www.stats.ox.ac.uk/ winkel/phs.html Dr Matthias Winkel 1 Outline I. z test for unknown population mean - review II. Limitations of the z test III. t test for unknown population

More information

Lecture Topic 6: Chapter 9 Hypothesis Testing

Lecture Topic 6: Chapter 9 Hypothesis Testing Lecture Topic 6: Chapter 9 Hypothesis Testing 9.1 Developing Null and Alternative Hypotheses Hypothesis testing can be used to determine whether a statement about the value of a population parameter should

More information

Null Hypothesis Significance Testing Signifcance Level, Power, t-tests Spring 2014 Jeremy Orloff and Jonathan Bloom

Null Hypothesis Significance Testing Signifcance Level, Power, t-tests Spring 2014 Jeremy Orloff and Jonathan Bloom Null Hypothesis Significance Testing Signifcance Level, Power, t-tests 18.05 Spring 2014 Jeremy Orloff and Jonathan Bloom Simple and composite hypotheses Simple hypothesis: the sampling distribution is

More information

Statistics - Written Examination MEC Students - BOVISA

Statistics - Written Examination MEC Students - BOVISA Statistics - Written Examination MEC Students - BOVISA Prof.ssa A. Guglielmi 26.0.2 All rights reserved. Legal action will be taken against infringement. Reproduction is prohibited without prior consent.

More information

Chi-Square & F Distributions

Chi-Square & F Distributions Chi-Square & F Distributions Carolyn J. Anderson EdPsych 580 Fall 2005 Chi-Square & F Distributions p. 1/55 Chi-Square & F Distributions... and Inferences about Variances The Chi-square Distribution Definition,

More information

STA218 Introduction to Hypothesis Testing

STA218 Introduction to Hypothesis Testing STA218 Introduction to Hypothesis Testing Al Nosedal. University of Toronto. Fall 2015 October 29, 2015 Who wants to be a millionaire? Let s say that one of you is invited to this popular show. As you

More information

Lecture 7: Binomial Test, Chisquare

Lecture 7: Binomial Test, Chisquare Lecture 7: Binomial Test, Chisquare Test, and ANOVA May, 01 GENOME 560, Spring 01 Goals ANOVA Binomial test Chi square test Fisher s exact test Su In Lee, CSE & GS suinlee@uw.edu 1 Whirlwind Tour of One/Two

More information

One-sample normal hypothesis Testing, paired t-test, two-sample normal inference, normal probability plots

One-sample normal hypothesis Testing, paired t-test, two-sample normal inference, normal probability plots 1 / 27 One-sample normal hypothesis Testing, paired t-test, two-sample normal inference, normal probability plots Timothy Hanson Department of Statistics, University of South Carolina Stat 704: Data Analysis

More information

The calculations lead to the following values: d 2 = 46, n = 8, s d 2 = 4, s d = 2, SEof d = s d n s d n

The calculations lead to the following values: d 2 = 46, n = 8, s d 2 = 4, s d = 2, SEof d = s d n s d n EXAMPLE 1: Paired t-test and t-interval DBP Readings by Two Devices The diastolic blood pressures (DBP) of 8 patients were determined using two techniques: the standard method used by medical personnel

More information

MAT X Hypothesis Testing - Part I

MAT X Hypothesis Testing - Part I MAT 2379 3X Hypothesis Testing - Part I Definition : A hypothesis is a conjecture concerning a value of a population parameter (or the shape of the population). The hypothesis will be tested by evaluating

More information

Hypothesis Testing. Chapter 7

Hypothesis Testing. Chapter 7 Hypothesis Testing Chapter 7 Hypothesis Testing Time to make the educated guess after answering: What the population is, how to extract the sample, what characteristics to measure in the sample, After

More information

Chapter 9: Comparison of Paired Samples

Chapter 9: Comparison of Paired Samples STAT503 Lecture Notes: Chapter 9 1 Chapter 9: Comparison of Paired Samples November 2, 2009 9.1 Introduction Observations occur in pairs such as: as identical twins, two observations on the same individual

More information

Lecture 13: Kolmogorov Smirnov Test & Power of Tests

Lecture 13: Kolmogorov Smirnov Test & Power of Tests Lecture 13: Kolmogorov Smirnov Test & Power of Tests S. Massa, Department of Statistics, University of Oxford 2 February 2016 An example Suppose you are given the following 100 observations. -0.16-0.68-0.32-0.85

More information

Structure of the Data. Paired Samples. Overview. The data from a paired design can be tabulated in this form. Individual Y 1 Y 2 d i = Y 1 Y

Structure of the Data. Paired Samples. Overview. The data from a paired design can be tabulated in this form. Individual Y 1 Y 2 d i = Y 1 Y Structure of the Data Paired Samples Bret Larget Departments of Botany and of Statistics University of Wisconsin Madison Statistics 371 11th November 2005 The data from a paired design can be tabulated

More information

Chapter 14: 1-6, 9, 12; Chapter 15: 8 Solutions When is it appropriate to use the normal approximation to the binomial distribution?

Chapter 14: 1-6, 9, 12; Chapter 15: 8 Solutions When is it appropriate to use the normal approximation to the binomial distribution? Chapter 14: 1-6, 9, 1; Chapter 15: 8 Solutions 14-1 When is it appropriate to use the normal approximation to the binomial distribution? The usual recommendation is that the approximation is good if np

More information

Testing: is my coin fair?

Testing: is my coin fair? Testing: is my coin fair? Formally: we want to make some inference about P(head) Try it: toss coin several times (say 7 times) Assume that it is fair ( P(head)= ), and see if this assumption is compatible

More information

An example ANOVA situation. 1-Way ANOVA. Some notation for ANOVA. Are these differences significant? Example (Treating Blisters)

An example ANOVA situation. 1-Way ANOVA. Some notation for ANOVA. Are these differences significant? Example (Treating Blisters) An example ANOVA situation Example (Treating Blisters) 1-Way ANOVA MATH 143 Department of Mathematics and Statistics Calvin College Subjects: 25 patients with blisters Treatments: Treatment A, Treatment

More information

MCQ TESTING OF HYPOTHESIS

MCQ TESTING OF HYPOTHESIS MCQ TESTING OF HYPOTHESIS MCQ 13.1 A statement about a population developed for the purpose of testing is called: (a) Hypothesis (b) Hypothesis testing (c) Level of significance (d) Test-statistic MCQ

More information

Hypothesis Testing Level I Quantitative Methods. IFT Notes for the CFA exam

Hypothesis Testing Level I Quantitative Methods. IFT Notes for the CFA exam Hypothesis Testing 2014 Level I Quantitative Methods IFT Notes for the CFA exam Contents 1. Introduction... 3 2. Hypothesis Testing... 3 3. Hypothesis Tests Concerning the Mean... 10 4. Hypothesis Tests

More information

I. Basics of Hypothesis Testing

I. Basics of Hypothesis Testing Introduction to Hypothesis Testing This deals with an issue highly similar to what we did in the previous chapter. In that chapter we used sample information to make inferences about the range of possibilities

More information

Power and Sample Size Determination

Power and Sample Size Determination Power and Sample Size Determination Bret Hanlon and Bret Larget Department of Statistics University of Wisconsin Madison November 3 8, 2011 Power 1 / 31 Experimental Design To this point in the semester,

More information

Exam 3 - Math Solutions - Spring Name : ID :

Exam 3 - Math Solutions - Spring Name : ID : Exam 3 - Math 3200 Solutions - Spring 2013 Name : ID : General instructions: This exam has 14 questions, each worth the same amount. Check that no pages are missing and promptly notify one of the proctors

More information

Hypothesis Testing with One Sample. Introduction to Hypothesis Testing 7.1. Hypothesis Tests. Chapter 7

Hypothesis Testing with One Sample. Introduction to Hypothesis Testing 7.1. Hypothesis Tests. Chapter 7 Chapter 7 Hypothesis Testing with One Sample 71 Introduction to Hypothesis Testing Hypothesis Tests A hypothesis test is a process that uses sample statistics to test a claim about the value of a population

More information

Sociology 6Z03 Topic 15: Statistical Inference for Means

Sociology 6Z03 Topic 15: Statistical Inference for Means Sociology 6Z03 Topic 15: Statistical Inference for Means John Fox McMaster University Fall 2016 John Fox (McMaster University) Soc 6Z03: Statistical Inference for Means Fall 2016 1 / 41 Outline: Statistical

More information

Tutorial 5: Hypothesis Testing

Tutorial 5: Hypothesis Testing Tutorial 5: Hypothesis Testing Rob Nicholls nicholls@mrc-lmb.cam.ac.uk MRC LMB Statistics Course 2014 Contents 1 Introduction................................ 1 2 Testing distributional assumptions....................

More information

Hypothesis Testing. Concept of Hypothesis Testing

Hypothesis Testing. Concept of Hypothesis Testing Quantitative Methods 2013 Hypothesis Testing with One Sample 1 Concept of Hypothesis Testing Testing Hypotheses is another way to deal with the problem of making a statement about an unknown population

More information

Classical Hypothesis Testing and GWAS

Classical Hypothesis Testing and GWAS Department of Computer Science Brown University, Providence sorin@cs.brown.edu September 15, 2014 Outline General Principles Classical statistical hypothesis testing involves the test of a null hypothesis

More information

Hypothesis Testing - II

Hypothesis Testing - II -3σ -2σ +σ +2σ +3σ Hypothesis Testing - II Lecture 9 0909.400.01 / 0909.400.02 Dr. P. s Clinic Consultant Module in Probability & Statistics in Engineering Today in P&S -3σ -2σ +σ +2σ +3σ Review: Hypothesis

More information

Statistics for Management II-STAT 362-Final Review

Statistics for Management II-STAT 362-Final Review Statistics for Management II-STAT 362-Final Review Multiple Choice Identify the letter of the choice that best completes the statement or answers the question. 1. The ability of an interval estimate to

More information

Inference in Regression Analysis. Dr. Frank Wood

Inference in Regression Analysis. Dr. Frank Wood Inference in Regression Analysis Dr. Frank Wood Inference in the Normal Error Regression Model Y i = β 0 + β 1 X i + ɛ i Y i value of the response variable in the i th trial β 0 and β 1 are parameters

More information

Nonparametric Methods for Two Samples. Nonparametric Methods for Two Samples

Nonparametric Methods for Two Samples. Nonparametric Methods for Two Samples Nonparametric Methods for Two Samples An overview In the independent two-sample t-test, we assume normality, independence, and equal variances. This t-test is robust against nonnormality, but is sensitive

More information

Math 62 Statistics Sample Exam Questions

Math 62 Statistics Sample Exam Questions Math 62 Statistics Sample Exam Questions 1. (10) Explain the difference between the distribution of a population and the sampling distribution of a statistic, such as the mean, of a sample randomly selected

More information

Statistics 104: Section 7

Statistics 104: Section 7 Statistics 104: Section 7 Section Overview Reminders Comments on Midterm Common Mistakes on Problem Set 6 Statistical Week in Review Comments on Midterm Overall, the midterms were good with one notable

More information

A) to D) to B) to E) to C) to Rate of hay fever per 1000 population for people under 25

A) to D) to B) to E) to C) to Rate of hay fever per 1000 population for people under 25 Unit 0 Review #3 Name: Date:. Suppose a random sample of 380 married couples found that 54 had two or more personality preferences in common. In another random sample of 573 married couples, it was found

More information

Homework Zero. Max H. Farrell Chicago Booth BUS41100 Applied Regression Analysis. Complete before the first class, do not turn in

Homework Zero. Max H. Farrell Chicago Booth BUS41100 Applied Regression Analysis. Complete before the first class, do not turn in Homework Zero Max H. Farrell Chicago Booth BUS41100 Applied Regression Analysis Complete before the first class, do not turn in This homework is intended as a self test of your knowledge of the basic statistical

More information

How to Conduct a Hypothesis Test

How to Conduct a Hypothesis Test How to Conduct a Hypothesis Test The idea of hypothesis testing is relatively straightforward. In various studies we observe certain events. We must ask, is the event due to chance alone, or is there some

More information

Chi-Square Tests. In This Chapter BONUS CHAPTER

Chi-Square Tests. In This Chapter BONUS CHAPTER BONUS CHAPTER Chi-Square Tests In the previous chapters, we explored the wonderful world of hypothesis testing as we compared means and proportions of one, two, three, and more populations, making an educated

More information

t-statistics for Weighted Means with Application to Risk Factor Models

t-statistics for Weighted Means with Application to Risk Factor Models t-statistics for Weighted Means with Application to Risk Factor Models Lisa R. Goldberg Barra, Inc. 00 Milvia St. Berkeley, CA 94704 Alec N. Kercheval Dept. of Mathematics Florida State University Tallahassee,

More information

Statistiek I. t-tests. John Nerbonne. CLCG, Rijksuniversiteit Groningen. John Nerbonne 1/35

Statistiek I. t-tests. John Nerbonne. CLCG, Rijksuniversiteit Groningen.  John Nerbonne 1/35 Statistiek I t-tests John Nerbonne CLCG, Rijksuniversiteit Groningen http://wwwletrugnl/nerbonne/teach/statistiek-i/ John Nerbonne 1/35 t-tests To test an average or pair of averages when σ is known, we

More information

Formulas and Tables by Mario F. Triola

Formulas and Tables by Mario F. Triola Formulas and Tables by Mario F. Triola Copyright 010 Pearson Education, Inc. Ch. 3: Descriptive Statistics x Mean f # x x f Mean (frequency table) 1x - x s B n - 1 Standard deviation n 1 x - 1 x Standard

More information

PROBLEM SET 1. For the first three answer true or false and explain your answer. A picture is often helpful.

PROBLEM SET 1. For the first three answer true or false and explain your answer. A picture is often helpful. PROBLEM SET 1 For the first three answer true or false and explain your answer. A picture is often helpful. 1. Suppose the significance level of a hypothesis test is α=0.05. If the p-value of the test

More information

Hypothesis testing: Examples. AMS7, Spring 2012

Hypothesis testing: Examples. AMS7, Spring 2012 Hypothesis testing: Examples AMS7, Spring 2012 Example 1: Testing a Claim about a Proportion Sect. 7.3, # 2: Survey of Drinking: In a Gallup survey, 1087 randomly selected adults were asked whether they

More information

Chapter 2 Probability Topics SPSS T tests

Chapter 2 Probability Topics SPSS T tests Chapter 2 Probability Topics SPSS T tests Data file used: gss.sav In the lecture about chapter 2, only the One-Sample T test has been explained. In this handout, we also give the SPSS methods to perform

More information

Lecture 13: Some Important Continuous Probability Distributions (Part 2)

Lecture 13: Some Important Continuous Probability Distributions (Part 2) Lecture 13: Some Important Continuous Probability Distributions (Part 2) Kateřina Staňková Statistics (MAT1003) May 14, 2012 Outline 1 Erlang distribution Formulation Application of Erlang distribution

More information

Practice Set 8b ST Spring 2014

Practice Set 8b ST Spring 2014 Practice Set 8b ST 20 - Spring 204 The following problem set will help you work with hypothesis testing.. Ex 8.7 (p. 400. Add the following parts: (c Explain why the assumption that interpupillary distance

More information

Comparing Two Populations OPRE 6301

Comparing Two Populations OPRE 6301 Comparing Two Populations OPRE 6301 Introduction... In many applications, we are interested in hypotheses concerning differences between the means of two populations. For example, we may wish to decide

More information

Biostatistics Lab Notes

Biostatistics Lab Notes Biostatistics Lab Notes Page 1 Lab 1: Measurement and Sampling Biostatistics Lab Notes Because we used a chance mechanism to select our sample, each sample will differ. My data set (GerstmanB.sav), looks

More information

fifty Fathoms Statistics Demonstrations for Deeper Understanding Tim Erickson

fifty Fathoms Statistics Demonstrations for Deeper Understanding Tim Erickson fifty Fathoms Statistics Demonstrations for Deeper Understanding Tim Erickson Contents What Are These Demos About? How to Use These Demos If This Is Your First Time Using Fathom Tutorial: An Extended Example

More information

Probability and Statistics Lecture 9: 1 and 2-Sample Estimation

Probability and Statistics Lecture 9: 1 and 2-Sample Estimation Probability and Statistics Lecture 9: 1 and -Sample Estimation to accompany Probability and Statistics for Engineers and Scientists Fatih Cavdur Introduction A statistic θ is said to be an unbiased estimator

More information

An Introduction to Bayesian Statistics

An Introduction to Bayesian Statistics An Introduction to Bayesian Statistics Robert Weiss Department of Biostatistics UCLA School of Public Health robweiss@ucla.edu April 2011 Robert Weiss (UCLA) An Introduction to Bayesian Statistics UCLA

More information

Chapter 11: Two Variable Regression Analysis

Chapter 11: Two Variable Regression Analysis Department of Mathematics Izmir University of Economics Week 14-15 2014-2015 In this chapter, we will focus on linear models and extend our analysis to relationships between variables, the definitions

More information

Do the following using Mintab (1) Make a normal probability plot for each of the two curing times.

Do the following using Mintab (1) Make a normal probability plot for each of the two curing times. SMAM 314 Computer Assignment 4 1. An experiment was performed to determine the effect of curing time on the comprehensive strength of concrete blocks. Two independent random samples of 14 blocks were prepared

More information

4.4. Further Analysis within ANOVA

4.4. Further Analysis within ANOVA 4.4. Further Analysis within ANOVA 1) Estimation of the effects Fixed effects model: α i = µ i µ is estimated by a i = ( x i x) if H 0 : µ 1 = µ 2 = = µ k is rejected. Random effects model: If H 0 : σa

More information

Seminar paper Statistics

Seminar paper Statistics Seminar paper Statistics The seminar paper must contain: - the title page - the characterization of the data (origin, reason why you have chosen this analysis,...) - the list of the data (in the table)

More information

Two-Variable Regression: Interval Estimation and Hypothesis Testing

Two-Variable Regression: Interval Estimation and Hypothesis Testing Two-Variable Regression: Interval Estimation and Hypothesis Testing Jamie Monogan University of Georgia Intermediate Political Methodology Jamie Monogan (UGA) Confidence Intervals & Hypothesis Testing

More information

2. The mean birthweights from this hospital is lower than 120 oz.

2. The mean birthweights from this hospital is lower than 120 oz. Probability February 13, 2013 Debdeep Pati Hypothesis testing Two sample inference Obstetrics: Mothers with low socioeconomic status (SES) deliver babies with lower than normal birthweights. The mean birthweights

More information

CHI-SQUARE TEST. The one-sample t test. Welch s two-sample t test in R. Welch s two-sample t test

CHI-SQUARE TEST. The one-sample t test. Welch s two-sample t test in R. Welch s two-sample t test The one-sample t test CHI-SQUARE TEST John Fry Boise State University One common statistic for hypothesis testing is the t statistic t = x µ s2 /N The t test looks at the mean x and variance s 2 of a sample

More information

Hypothesis Testing. Dr. Bob Gee Dean Scott Bonney Professor William G. Journigan American Meridian University

Hypothesis Testing. Dr. Bob Gee Dean Scott Bonney Professor William G. Journigan American Meridian University Hypothesis Testing Dr. Bob Gee Dean Scott Bonney Professor William G. Journigan American Meridian University 1 AMU / Bon-Tech, LLC, Journi-Tech Corporation Copyright 2015 Learning Objectives Upon successful

More information

Power & Effect Size power Effect Size

Power & Effect Size power Effect Size Power & Effect Size Until recently, researchers were primarily concerned with controlling Type I errors (i.e. finding a difference when one does not truly exist). Although it is important to make sure

More information

Module 9: Nonparametric Tests. The Applied Research Center

Module 9: Nonparametric Tests. The Applied Research Center Module 9: Nonparametric Tests The Applied Research Center Module 9 Overview } Nonparametric Tests } Parametric vs. Nonparametric Tests } Restrictions of Nonparametric Tests } One-Sample Chi-Square Test

More information

Inferential Statistics

Inferential Statistics Inferential Statistics Sampling and the normal distribution Z-scores Confidence levels and intervals Hypothesis testing Commonly used statistical methods Inferential Statistics Descriptive statistics are

More information

Chi Square for Contingency Tables

Chi Square for Contingency Tables 2 x 2 Case Chi Square for Contingency Tables A test for p 1 = p 2 We have learned a confidence interval for p 1 p 2, the difference in the population proportions. We want a hypothesis testing procedure

More information

3.6: General Hypothesis Tests

3.6: General Hypothesis Tests 3.6: General Hypothesis Tests The χ 2 goodness of fit tests which we introduced in the previous section were an example of a hypothesis test. In this section we now consider hypothesis tests more generally.

More information

Analysis of numerical data S4

Analysis of numerical data S4 Basic medical statistics for clinical and experimental research Analysis of numerical data S4 Katarzyna Jóźwiak k.jozwiak@nki.nl 3rd November 2015 1/42 Hypothesis tests: numerical and ordinal data 1 group:

More information

Data Analysis and Uncertainty Part 3: Hypothesis Testing/Sampling

Data Analysis and Uncertainty Part 3: Hypothesis Testing/Sampling Data Analysis and Uncertainty Part 3: Hypothesis Testing/Sampling Instructor: Sargur N. University at Buffalo The State University of New York srihari@cedar.buffalo.edu Topics 1. Hypothesis Testing 1.

More information

Statistiek (WISB361)

Statistiek (WISB361) Statistiek (WISB361) Final exam June 29, 2015 Schrijf uw naam op elk in te leveren vel. Schrijf ook uw studentnummer op blad 1. The maximum number of points is 100. Points distribution: 23 20 20 20 17

More information

Statistics II Final Exam - January Use the University stationery to give your answers to the following questions.

Statistics II Final Exam - January Use the University stationery to give your answers to the following questions. Statistics II Final Exam - January 2012 Use the University stationery to give your answers to the following questions. Do not forget to write down your name and class group in each page. Indicate clearly

More information

Statistics: revision

Statistics: revision NST 1B Experimental Psychology Statistics practical 5 Statistics: revision Rudolf Cardinal & Mike Aitken 3 / 4 May 2005 Department of Experimental Psychology University of Cambridge Slides at pobox.com/~rudolf/psychology

More information

Simple Linear Regression

Simple Linear Regression Inference for Regression Simple Linear Regression IPS Chapter 10.1 2009 W.H. Freeman and Company Objectives (IPS Chapter 10.1) Simple linear regression Statistical model for linear regression Estimating

More information

41.2. Tests Concerning asinglesample. Introduction. Prerequisites. Learning Outcomes

41.2. Tests Concerning asinglesample. Introduction. Prerequisites. Learning Outcomes Tests Concerning asinglesample 41.2 Introduction This Section introduces you to the basic ideas of hypothesis testing in a non-mathematical way by using a problem solving approach to highlight the concepts

More information

Nonparametric Test Procedures

Nonparametric Test Procedures Nonparametric Test Procedures 1 Introduction to Nonparametrics Nonparametric tests do not require that samples come from populations with normal distributions or any other specific distribution. Hence

More information

MAT 155. Key Concept. September 27, 2010. 155S5.5_3 Poisson Probability Distributions. Chapter 5 Probability Distributions

MAT 155. Key Concept. September 27, 2010. 155S5.5_3 Poisson Probability Distributions. Chapter 5 Probability Distributions MAT 155 Dr. Claude Moore Cape Fear Community College Chapter 5 Probability Distributions 5 1 Review and Preview 5 2 Random Variables 5 3 Binomial Probability Distributions 5 4 Mean, Variance and Standard

More information

m (t) = e nt m Y ( t) = e nt (pe t + q) n = (pe t e t + qe t ) n = (qe t + p) n

m (t) = e nt m Y ( t) = e nt (pe t + q) n = (pe t e t + qe t ) n = (qe t + p) n 1. For a discrete random variable Y, prove that E[aY + b] = ae[y] + b and V(aY + b) = a 2 V(Y). Solution: E[aY + b] = E[aY] + E[b] = ae[y] + b where each step follows from a theorem on expected value from

More information

Hypothesis tests, confidence intervals, and bootstrapping

Hypothesis tests, confidence intervals, and bootstrapping Hypothesis tests, confidence intervals, and bootstrapping Business Statistics 41000 Fall 2015 1 Topics 1. Hypothesis tests Testing a mean: H0 : µ = µ 0 Testing a proportion: H0 : p = p 0 Testing a difference

More information

c. The factor is the type of TV program that was watched. The treatment is the embedded commercials in the TV programs.

c. The factor is the type of TV program that was watched. The treatment is the embedded commercials in the TV programs. STAT E-150 - Statistical Methods Assignment 9 Solutions Exercises 12.8, 12.13, 12.75 For each test: Include appropriate graphs to see that the conditions are met. Use Tukey's Honestly Significant Difference

More information

Confidence Intervals

Confidence Intervals Confidence Intervals List of things to Do:- Compute the point estimate of µ Compute the confidence intervals about µ with σ known Error term in the confidence interval Determine the sample size n for estimating

More information

Math 10 MPS Homework 6 Answers to additional problems

Math 10 MPS Homework 6 Answers to additional problems Math 1 MPS Homework 6 Answers to additional problems 1. What are the two types of hypotheses used in a hypothesis test? How are they related? Ho: Null Hypotheses A statement about a population parameter

More information

Lecture 4c: Practice Problems. John McGready Johns Hopkins University

Lecture 4c: Practice Problems. John McGready Johns Hopkins University This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike License. Your use of this material constitutes acceptance of that license and the conditions of use of materials on this

More information

MATH4427 Notebook 3 Spring MATH4427 Notebook Hypotheses: Definitions and Examples... 3

MATH4427 Notebook 3 Spring MATH4427 Notebook Hypotheses: Definitions and Examples... 3 MATH4427 Notebook 3 Spring 2016 prepared by Professor Jenny Baglivo c Copyright 2009-2016 by Jenny A. Baglivo. All Rights Reserved. Contents 3 MATH4427 Notebook 3 3 3.1 Hypotheses: Definitions and Examples............................

More information

Chi-Square Test. Contingency Tables. Contingency Tables. Chi-Square Test for Independence. Chi-Square Tests for Goodnessof-Fit

Chi-Square Test. Contingency Tables. Contingency Tables. Chi-Square Test for Independence. Chi-Square Tests for Goodnessof-Fit Chi-Square Tests 15 Chapter Chi-Square Test for Independence Chi-Square Tests for Goodness Uniform Goodness- Poisson Goodness- Goodness Test ECDF Tests (Optional) McGraw-Hill/Irwin Copyright 2009 by The

More information

Hypothesis Testing hypothesis testing approach formulation of the test statistic

Hypothesis Testing hypothesis testing approach formulation of the test statistic Hypothesis Testing For the next few lectures, we re going to look at various test statistics that are formulated to allow us to test hypotheses in a variety of contexts: In all cases, the hypothesis testing

More information

Statistical foundations of machine learning

Statistical foundations of machine learning Machine learning p. 1/45 Statistical foundations of machine learning INFO-F-422 Gianluca Bontempi Département d Informatique Boulevard de Triomphe - CP 212 http://www.ulb.ac.be/di Machine learning p. 2/45

More information

6.1 The Elements of a Test of Hypothesis

6.1 The Elements of a Test of Hypothesis University of California, Davis Department of Statistics Summer Session II Statistics 13 August 22, 2012 Date of latest update: August 20 Lecture 6: Tests of Hypothesis Suppose you wanted to determine

More information

Hypothesis Testing. April 21, 2009

Hypothesis Testing. April 21, 2009 Hypothesis Testing April 21, 2009 Your Claim is Just a Hypothesis I ve never made a mistake. Once I thought I did, but I was wrong. Your Claim is Just a Hypothesis Confidence intervals quantify how sure

More information

Statistics revision. Dr. Inna Namestnikova. Statistics revision p. 1/8

Statistics revision. Dr. Inna Namestnikova. Statistics revision p. 1/8 Statistics revision Dr. Inna Namestnikova inna.namestnikova@brunel.ac.uk Statistics revision p. 1/8 Introduction Statistics is the science of collecting, analyzing and drawing conclusions from data. Statistics

More information

Today s lecture. Lecture 6: Dichotomous Variables & Chi-square tests. Proportions and 2 2 tables. How do we compare these proportions

Today s lecture. Lecture 6: Dichotomous Variables & Chi-square tests. Proportions and 2 2 tables. How do we compare these proportions Today s lecture Lecture 6: Dichotomous Variables & Chi-square tests Sandy Eckel seckel@jhsph.edu Dichotomous Variables Comparing two proportions using 2 2 tables Study Designs Relative risks, odds ratios

More information

The t Test. April 3-8, exp (2πσ 2 ) 2σ 2. µ ln L(µ, σ2 x) = 1 σ 2. i=1. 2σ (σ 2 ) 2. ˆσ 2 0 = 1 n. (x i µ 0 ) 2 = i=1

The t Test. April 3-8, exp (2πσ 2 ) 2σ 2. µ ln L(µ, σ2 x) = 1 σ 2. i=1. 2σ (σ 2 ) 2. ˆσ 2 0 = 1 n. (x i µ 0 ) 2 = i=1 The t Test April 3-8, 008 Again, we begin with independent normal observations X, X n with unknown mean µ and unknown variance σ The likelihood function Lµ, σ x) exp πσ ) n/ σ x i µ) Thus, ˆµ x ln Lµ,

More information

Linear constraints in multiple linear regression. Analysis of variance.

Linear constraints in multiple linear regression. Analysis of variance. Section 16 Linear constraints in multiple linear regression. Analysis of variance. Multiple linear regression with general linear constraints. Let us consider a multiple linear regression Y = X + β and

More information

1. Chi-Squared Tests

1. Chi-Squared Tests 1. Chi-Squared Tests We'll now look at how to test statistical hypotheses concerning nominal data, and specifically when nominal data are summarized as tables of frequencies. The tests we will considered

More information

Hypothesis Testing COMP 245 STATISTICS. Dr N A Heard. 1 Hypothesis Testing 2 1.1 Introduction... 2 1.2 Error Rates and Power of a Test...

Hypothesis Testing COMP 245 STATISTICS. Dr N A Heard. 1 Hypothesis Testing 2 1.1 Introduction... 2 1.2 Error Rates and Power of a Test... Hypothesis Testing COMP 45 STATISTICS Dr N A Heard Contents 1 Hypothesis Testing 1.1 Introduction........................................ 1. Error Rates and Power of a Test.............................

More information

Stat 371, Cecile Ane Practice problems Midterm #2, Spring 2012

Stat 371, Cecile Ane Practice problems Midterm #2, Spring 2012 Stat 371, Cecile Ane Practice problems Midterm #2, Spring 2012 The first 3 problems are taken from previous semesters exams, with solutions at the end of this document. The other problems are suggested

More information

Lecture 5 : The Poisson Distribution. Jonathan Marchini

Lecture 5 : The Poisson Distribution. Jonathan Marchini Lecture 5 : The Poisson Distribution Jonathan Marchini Random events in time and space Many experimental situations occur in which we observe the counts of events within a set unit of time, area, volume,

More information

Statistical Significance and Bivariate Tests

Statistical Significance and Bivariate Tests Statistical Significance and Bivariate Tests BUS 735: Business Decision Making and Research 1 1.1 Goals Goals Specific goals: Re-familiarize ourselves with basic statistics ideas: sampling distributions,

More information

Example 11-3, pg. 495 The Chi-Square Goodness-of-Fit Test

Example 11-3, pg. 495 The Chi-Square Goodness-of-Fit Test 132 Chapter 11 Chi-Square Tests Chi-Square Tests Chapter 11 Section 11.2 Example 11-3, pg. 495 The Chi-Square Goodness-of-Fit Test A bank manager wanted to investigate if the percentage of people who use

More information

Math 1070 Exam 2B 22 March, 2013

Math 1070 Exam 2B 22 March, 2013 Math 1070 Exam 2B 22 March, 2013 This exam will last 50 minutes and consists of 13 multiple choice and 6 free response problems. Write your answers in the space provided. All solutions must be sufficiently

More information

Tukey s HSD (Honestly Significant Difference).

Tukey s HSD (Honestly Significant Difference). Agenda for Week 4 (Tuesday, Jan 26) Week 4 Hour 1 AnOVa review. Week 4 Hour 2 Multiple Testing Tukey s HSD (Honestly Significant Difference). Week 4 Hour 3 (Thursday) Two-way AnOVa. Sometimes you ll need

More information

Inference for Regression

Inference for Regression Simple Linear Regression Inference for Regression The simple linear regression model Estimating regression parameters; Confidence intervals and significance tests for regression parameters Inference about

More information