How Big Should Sample Size be? Example We have data y=(y 1,…,y N ), where y~N(μ,σ 2 ) We want to test H 0 : μ=θ vs H 1 : μ θ –Chosen significance level=α=0.01.

Slides:



Advertisements
Similar presentations
Chart 11-1: Minitab Difference Between 2 Population Means
Advertisements

The t Test for Two Independent Samples
Introduction to Hypothesis Testing
Section 7.2. Mean of a probability distribution is the long- run average outcome, µ, or µ x. Also called the expected value of x, or E(X). µ x = x i P.
Slide 8- 1 Copyright © 2010, 2007, 2004 Pearson Education, Inc. All Rights Reserved. Active Learning Lecture Slides For use with Classroom Response Systems.
Introductory Mathematics & Statistics for Business
1 Superior Safety in Noninferiority Trials David R. Bristol To appear in Biometrical Journal, 2005.
Significance testing Ioannis Karagiannis (based on previous EPIET material) 18 th EPIET/EUPHEM Introductory course
Type I and Type II errors
Chapter 25 Paired Samples and Blocks
Chapter 17/18 Hypothesis Testing
Hypothesis Test II: t tests
Department of Engineering Management, Information and Systems
CHAPTER EIGHT TESTS OF HYPOTHESES
On Comparing Classifiers : Pitfalls to Avoid and Recommended Approach
One-sample T-Test Matched Pairs T-Test Two-sample T-Test
HYPOTHESIS TESTING. Purpose The purpose of hypothesis testing is to help the researcher or administrator in reaching a decision concerning a population.
(c) 2007 IUPUI SPEA K300 (4392) Outline Type of t-test Z-test versus t-test Assumptions of the t-test One sample t-test Paired sample t-test F-test for.
June 14. In Chapter 9: 9.1 Null and Alternative Hypotheses 9.2 Test Statistic 9.3 P-Value 9.4 Significance Level 9.5 One-Sample z Test 9.6 Power and Sample.
INTRODUCTORY STATISTICS FOR CRIMINAL JUSTICE
Hypothesis Testing with t Tests
7.2 Hypothesis Testing for the Mean (Large Samples)
Introduction to Hypothesis Testing
6. Statistical Inference: Example: Anorexia study Weight measured before and after period of treatment y i = weight at end – weight at beginning For n=17.
Sample size and statistical power ดร. อรพิน กฤษณเกรียงไกร ดร. อรพิน กฤษณเกรียงไกรคณะสาธารณสุขศาสตร์มหาวิทยาลัยนเรศวร.
Minimizing Chance of Type I and Type II Errors
Power and sample size.
There are two statistical tests for mean: 1) z test – Used for large samples (n ≥ 30) 1) t test – Used for small samples (n < 30)
T-Tests For Dummies As in the books, not you personally!
“Students” t-test.
Comparing Two Population Parameters
Final Review Answers. Two-Tailed t Test Solution -1 H 0 : H a :  = df = Critical Value(s): Test Statistic: Decision:  = 368   = 24 t.
Module 16: One-sample t-tests and Confidence Intervals
Module 17: Two-Sample t-tests, with equal variances for the two populations This module describes one of the most utilized statistical tests, the.
Chapter 8: Introduction to Hypothesis Testing. 2 Hypothesis Testing An inferential procedure that uses sample data to evaluate the credibility of a hypothesis.
T- Test Example of how to perform a t-test. Step 1 State the null hypothesis  Begin with assuming that any observed differences between the two samples.
Chapter 8 test Test on Chapter 8 will be on: Feb 28, 2013 Bluman, Chapter 81.
McGraw-Hill, Bluman, 7th ed., Chapter 9
Please enter data on page 477 in your calculator.
Using t-tests Basic introduction and 1-sample t-tests Statistics for the Social Sciences Psychology 340 Spring 2010.
Statistics for the Social Sciences
Statistics for the Social Sciences
Statistical Inferences Based on Two Samples
1 Copyright © 2005 Brooks/Cole, a division of Thomson Learning, Inc. Two Sample Proportions Large Sample Difference of Proportions z Test & Confidence.
Analysis of Variance Chapter 12 . McGraw-Hill/Irwin
Multiple Regression and Model Building
Chapter 8 Hypothesis Test.
STATISTICAL INFERENCE PART IV LOCATION AND SCALE PARAMETERS 1.
Chapter 13 Comparing Two Populations: Independent Samples.
MATH 4030 – 11A ANALYSIS OF R X C TABLES (GOODNESS-OF-FIT TEST) 1.
Week 101 Test on Pairs of Means – Case I Suppose are iid independent of that are iid. Further, suppose that n 1 and n 2 are large or that are known. We.
Simulation – Output Analysis. Point Estimate Point estimates are single value estimates of parameters of interest (in this case the mean and standard.
Unit 8 Section 7.4.
Making Sense of Statistical Significance Inference as Decision
Chapter 9 Hypothesis Testing.
Unit 8 Section 7.5.
Power and Effect Size.
Statistics for the Social Sciences
Hypothesis Testing II: The Two-sample Case
Statistics Chapter 7 Review.
Confidence Intervals and Hypothesis Tests for Variances for One Sample
Chapter 8 Hypothesis Testing with Two Samples.
Understanding How Hypothesis Testing Works
Hypothesis Tests for a Population Mean in Practice
Estimates Made Using Sx
Inference about Two Means: Independent Samples
Lesson Use and Abuse of Tests.
Elementary Statistics: Picturing The World
SMOKERS NONSMOKERS Sample 1, size n1 Sample 2, size n2
Graphs and Chi Square.
Presentation transcript:

How Big Should Sample Size be? Example We have data y=(y 1,…,y N ), where y~N(μ,σ 2 ) We want to test H 0 : μ=θ vs H 1 : μ θ –Chosen significance level=α=0.01 Fixed critical cut-off value = c * = 1.64, –Test statistics –If |z|>1,64, reject H0

How Big Should Sample Size be? (contd) An experiment collected N=100 observations, y N=100, assumed to be independently from N(µ,1) He wants to test H 0 : µ=0 vs H 1 : µ0 Current sample mean y N=100 =0.2. So, z N=100 =(0.2-0)*10=2 Pocock (1977) showed cut-off value under α=.05, c*=2.18 * Since z100<c, H0 is not rejected. So, he decided to collect another 100 observations. Then, he computed z N=200 ………….. Could he reject H0 eventually?

Business Concept Testing To evaluate feasibility of a business idea Two key decisions 1. How to communicate the concept 2. The data to collect from respondents Respondents –Potential target customers –Other important players (e.g., possible channel members)

Measures –Primary Intended purchase measures Overall product diagnostics Special attribute diagnostics Respondents profiling variables –Secondary Open-ended reason why

How to Present A Business Concept? Use simple written statement on the business concept (possibly with visual stimuli) –A short statement of the core product concept –Alternatively, a vivid story board

Your Task Develop a research plan for concept testing of LINEPASS. Objective Respondents Sample size Method Stimuli Data Measurement Analysis

Affinity Diagram (Kawkita Jiro Method)

Fishbone Diagram

Market Entry Strategies These firms… are characterized by these elements: Successful pioneers [ Netscape] -Large entry scale -Broad product line -High product quality Successful fast followers [ Microsoft ] -Larger entry scale than the pioneer -Leapfrogging the pioneer with superior technology, quality and customer service. Successful late entrant -Focus on peripheral target markets or niches Mass-Market Penetration Niche Penetration

Market Entry Strategy for Pioneer 1.Mass Penetration Netscape 2.Niche Penetration 3.Focused Sequential Penetration Trakus

Situations Favoring Alternative Marketing Strategies for Pioneers Mass-Market PenetrationFocused Sequential Penetration Market Characteristics-large potential demand -homogeneous customer needs -short diffusion process -large potential demand -fragmented market -relatively longer diffusion process Product Characteristics-difficult to copy -limited sources of supply -complex production process -easily copied or adopted -many sources of supply -relatively simple production process Competitor Characteristics-few potential competitors -Few sources of differential advantage -many potential competitors -many possible sources of differential advantage Firm Characteristics-strong product engineering skills -strong marketing skills and resources -sufficient financial and organizational resources -limited product engineering skills and resources -limited marketing skills and resources -insufficient financial and organizational resources

Strategic Objectives Mass-Market PenetrationFocused Penetration Short-term-Maximize number of adopters in total market -Invest heavily to build future volume and share -Maximize number of adopters in initial target segment -Limited investment to build volume and share in chosen initial niche Intermediate-term-Attempt to preempt competition -Maintain leading market position even if some sacrifice of margins is necessary in short terms as new competitors enter -Maintain leading share position in target segment even if some sacrifice of margins is necessary in short terms as new competitors enter -Accumulate resources and experience from initial niche and utilize them to penetrate other segments. Long-term-Maximize ROI WildFire

EPILOGUE In January 1995, McCaw invested $5 million in Wildfire and obtained 12.5% of the companys shares –Mutually non-exclusive collaboration for Network Wildfire –McCaw placed an order for a 500-user prototype –Planned to scale up a 500-user prototype Network Wildfire to a 1,500 user system by the end of In July 2000, Wildfire was acquired for about $142 million by Orange Telecommunications, a subsidiary of France Telecom. Currently, Wildfire is focusing on Network and Corporate systems.

Entrepreneurs do not have sufficient information –Lots of missing information –many hypothesis However, it is crucial to make on-line correction of initial plan over time. Entrepreneurs need to gather critical information in order to make correct on-line correction of plan. –Fill missing information –Test hypothesis Wildfires decision on line-of-business highlights the need of obtaining critical information for its on-line correction of plan. Wildfire case also highlights the importance of partnership in order to overcome the problem of limited resources.