Week 11 Lecture Slides.pdf - OIS 2340 Business Statistics Ben Helland MBA Fall 2018 Tonight\u2019s Agenda \u25a1HW 7 Review \u25a1Ch 9 \u2013 Null-busters The Road

# Week 11 Lecture Slides.pdf - OIS 2340 Business Statistics...

This preview shows page 1 - 11 out of 48 pages.

OIS 2340Business StatisticsBen Helland, MBA Fall 2018
Tonight’s AgendaHW 7 Review Ch 9 – Null-busters
The Road MapAre we there yet?Definitions"Middle"VariationProbabilityStandardiz-ationInferenceHypothesis Testing
Introduction to Hypothesis Testing - MeansOne of these things is not like the other
Hypothesis Testing: The EndWhy do we do hypothesis testing:Structured analytical method for making decisions that account for sampling errorRecall that sampling error = difference between sample mean and population meanProbability of decision’s errors are controlled or at least control the level of uncertainty“Beyond reasonable doubt” vs “innocent until proven guilty”H0
Hypothesis Testing: The EndDefinitions and terms –Null hypothesis: H0Statement about the population parameter that will be assumed to be trueduring the hypothesis test. What you test against!**Can only be rejected or not rejected**Alternative hypothesis: HAHypothesis that includes all population values not included in the Null Hypothesis. Only selected if there is strong enoughsample evidence to reject the Null Hypothesis. What you’re trying to prove!Will be tested via equality signs:“<=“, “>=”, ”=“THIS IS THE TRICKIEST PART -identifying what the null and alt hypothesisare/meanH0
Hypothesis Testing: The EndTypes of testing:Testing the status quoHas something changed? (worse, better, not the same?)Testing a research questionWe’ve done something, is it significant?Testing a claim about the populationBurden of proof on the claim (about population)If claim contains an equality it is the null hypothesisH0
Hypothesis Testing: The EndFormulating a null and alt hypothesis: Status Quo Examples:You manage a LEGO factory manufacturing line. You know the ideal place to have your tolerances on a certain piece is 400mm. You pull a sample of the line and test whether or not that sample is significantly different from what you intend. H0
Hypothesis Testing: The EndFormulating a null and alt hypothesis: Status Quo Examples:You manage a LEGO factory manufacturing line. You know the ideal place to have your tolerances on a certain piece is 400mm. You pull a sample of the line and test whether or not that sample is significantly different from what you intend. H0: μ = 400mmHA: μ 400mmNo direction stated…H0
Hypothesis Testing: The EndFormulating a null and alt hypothesis: Status Quo Examples:UDOT has local mandates that intersections must have wait times no longer than 5 mins. Occasionally the sample intersections’ red lights and time how long the light took to change, in order to assess whether or not the times are longer than guidelines require.

#### You've reached the end of your free preview.

Want to read all 48 pages?

• Fall '11
• Null hypothesis, Statistical hypothesis testing, Type I and type II errors