(mu, pronounced mew ) and standard deviation σ (sigma).

Similar documents
Today - SPSS and standard error - End of Midterm 1 exam material - T-scores

Alex s Guide to Word Problems and Linear Equations Following Glencoe Algebra 1

Note that we are looking at the true mean, μ, not y. The problem for us is that we need to find the endpoints of our interval (a, b).

Position and Displacement

Note that we are looking at the true mean, μ, not y. The problem for us is that we need to find the endpoints of our interval (a, b).

Math 31 Lesson Plan. Day 2: Sets; Binary Operations. Elizabeth Gillaspy. September 23, 2011

MA 1125 Lecture 15 - The Standard Normal Distribution. Friday, October 6, Objectives: Introduce the standard normal distribution and table.

Wed, June 26, (Lecture 8-2). Nonlinearity. Significance test for correlation R-squared, SSE, and SST. Correlation in SPSS.

Direct Proof MAT231. Fall Transition to Higher Mathematics. MAT231 (Transition to Higher Math) Direct Proof Fall / 24

Confidence Intervals. - simply, an interval for which we have a certain confidence.

Math 31 Lesson Plan. Day 16: Review; Start Section 8. Elizabeth Gillaspy. October 18, Supplies needed: homework. Colored chalk. Quizzes!

1 Review of the dot product

Algebra 1B notes and problems March 12, 2009 Factoring page 1

Stat 20 Midterm 1 Review

ADVENTURES IN THE FLIPPED CLASSROOM FOR INTRODUCTORY

Section 9 2B:!! Using Confidence Intervals to Estimate the Difference ( µ 1 µ 2 ) in Two Population Means using Two Independent Samples.

Chapter 6 The Normal Distribution

Math 31 Lesson Plan. Day 5: Intro to Groups. Elizabeth Gillaspy. September 28, 2011

Last few slides from last time

Business Statistics. Lecture 9: Simple Regression

Math 111, Introduction to the Calculus, Fall 2011 Midterm I Practice Exam 1 Solutions

Determining the Spread of a Distribution

Determining the Spread of a Distribution

Multiple Regression Analysis

Lesson 6-1: Relations and Functions

An inferential procedure to use sample data to understand a population Procedures

Math 3B: Lecture 11. Noah White. October 25, 2017

Mathematical Induction. EECS 203: Discrete Mathematics Lecture 11 Spring

Confidence intervals

Recitation 9: Probability Matrices and Real Symmetric Matrices. 3 Probability Matrices: Definitions and Examples

Math 163 (23) - Midterm Test 1

Statistics for IT Managers

Probability Distributions

CH 23 PRODUCING WIDGETS

9. TRANSFORMING TOOL #2 (the Multiplication Property of Equality)

Distribution-Free Procedures (Devore Chapter Fifteen)

PROBLEM SET 3: PROOF TECHNIQUES

Date Lesson Text TOPIC Homework. Verifying Equations WS 3.1. Solving Multi-Step Equations WS 3.3. Solving Equations with Rationals WS 3.

4.2 The Normal Distribution. that is, a graph of the measurement looks like the familiar symmetrical, bell-shaped

P (A) = P (B) = P (C) = P (D) =

λ λ λ In-class problems

1 Probability Distributions

Recap: Language models. Foundations of Natural Language Processing Lecture 4 Language Models: Evaluation and Smoothing. Two types of evaluation in NLP

Math WW08 Solutions November 19, 2008

Lecture 30. DATA 8 Summer Regression Inference

CS 154 Introduction to Automata and Complexity Theory

Stat 139 Homework 2 Solutions, Spring 2015

GUIDED NOTES 2.5 QUADRATIC EQUATIONS

Quantitative Understanding in Biology Module II: Model Parameter Estimation Lecture I: Linear Correlation and Regression

Math 112 Group Activity: The Vertical Speed of a Shell

Introduction to Linear Regression

Business Statistics Midterm Exam Fall 2015 Russell. Please sign here to acknowledge

Part III: A Simplex pivot

Unit 4 Day 4 & 5. Piecewise Functions

CHAPTER 5: EXPLORING DATA DISTRIBUTIONS. Individuals are the objects described by a set of data. These individuals may be people, animals or things.

X = X X n, + X 2

Pr[X = s Y = t] = Pr[X = s] Pr[Y = t]

N/4 + N/2 + N = 2N 2.

Section 1.2 Factors and Factor Operators

Equivalence Relations and Partitions, Normal Subgroups, Quotient Groups, and Homomorphisms

Physics 351 Wednesday, January 10, 2018

C if U can. Algebra. Name

31. TRANSFORMING TOOL #2 (the Multiplication Property of Equality)

Section 5.4. Ken Ueda

MATH 310, REVIEW SHEET 2

Going from graphic solutions to algebraic

Elementary Statistics

Distributive property and its connection to areas

/633 Introduction to Algorithms Lecturer: Michael Dinitz Topic: Asymptotic Analysis, recurrences Date: 9/7/17

Announcements Wednesday, August 30

Bias Variance Trade-off

Unit 4: Part 3 Solving Quadratics

Math 5a Reading Assignments for Sections

Acceleration 1-D Motion for Calculus Students (90 Minutes)

. Do the assigned problems on separate paper and show your work

Unit 8 - Polynomial and Rational Functions Classwork

Student Activity: Finding Factors and Prime Factors

Announcements Wednesday, August 30

Lecture 20: Further graphing

CH 58 PRODUCING WIDGETS

Steps to take to do the descriptive part of regression analysis:

Lecture 17: Floyd-Hoare Logic for Partial Correctness

1 Lesson 13: Methods of Integration

Astronomy 102 Math Review

(i) Write as a mixed number. (ii) Work out of (iii) Complete each statement with the correct symbol. = < >

Physics 351 Wednesday, January 14, 2015

Discrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 20

4. What verb is used to describe Earth s

Park School Mathematics Curriculum Book 9, Lesson 2: Introduction to Logarithms

2 Lecture Defining Optimization with Equality Constraints

Math 3C Lecture 20. John Douglas Moore

Loose Ends and Final Review

MATH CRASH COURSE GRA6020 SPRING 2012

8. TRANSFORMING TOOL #1 (the Addition Property of Equality)

SOLUTIONS Math 345 Homework 6 10/11/2017. Exercise 23. (a) Solve the following congruences: (i) x (mod 12) Answer. We have

HOW TO WRITE PROOFS. Dr. Min Ru, University of Houston

Dynamic Programming: Matrix chain multiplication (CLRS 15.2)

Solving Quadratic & Higher Degree Equations

Hi, my name is Dr. Ann Weaver of Argosy University. This WebEx is about something in statistics called z-

Transcription:

STAT 203 Lecture 4-1. - The normal distribution is symmetric. - Getting the probability from between two z-scores - Translating standard scores to and from raw scores. - Extreme values beyond the table. So Majestic!

Text from last Friday: Say a value X followed the normal distribution, with mean μ (mu, pronounced mew ) and standard deviation σ (sigma). We used the z-table to find things like the probability that X is greater than 1.28 standard deviations above the mean. In other words, we found Pr( X > μ + 1.28σ)

μ + 1.28σ means a z-score of 1.28. From the z-table, page 515 z Area between Mean Area beyond z and z 1.27 39.80 10.20 1.28 39.97 10.03 1.29 40.15 9.85 Since we re looking at the values farther away from the mean than the cutoff, we want the area beyond z.

Pr( X > μ + 1.28σ) = 10.03%, or about 10% Can we find Pr( X > μ - 1.28σ)? Hint: Think symmetry.

We can find Pr( X > μ - 1.28σ) Symmetry: The same on both sides.

We can find Pr( X > μ - 1.28σ) Symmetry: The same on both sides.

What is the chance that this value, X, is more than 1 standard deviation away the mean in either direction? Start with Pr( X > μ + 1σ), or, because it s simpler to write: Pr( Z > 1) By the table (page 514) z Area between Mean Area beyond z and z.99 33.89 16.11 1.00 34.13 15.87 1.01 34.38 15.62

Pr( Z > 1) =.16

Pr( Z > 1) =.16, so Pr( Z < -1) =.16 also

Pr( Z > 1) + Pr(Z < -1) =.32 Not surprizing since Pr( -1 < Z < 1) =.68,.68 +.32 = 1.00

We could have done this the other way too: Working backwards from Pr( -1 < Z < 1) =.68 We could get by converse Pr(Z < -1) + Pr(Z > 1) =.32 and get by symmetry Pr(Z > 1) =.16 One other thing to note is that Z = 0 right at the mean, because the mean is 0 standard deviations above or below the mean.

Let s try with some uglier z-scores. Pr( -1.75 < Z < 0.52) z Area between Mean Area beyond z and z 0.51 19.50 30.50 0.52 19.85 30.15 1.74 45.91 4.09 1.75 45.99 4.01

Doing the math

Pr( -1.75 < Z < 0.52) can be split into two ranges using the mean as the split point. Pr( -1.75 < Z < 0 ) + Pr( 0 < Z < 0.52) Why would we do this? Because the table has everything from the mean. Pr(-1.75 < Z < 0) =.4599 Pr(0 < Z < 0.52) =.1985.4599 +.1985 =.6584 About 66% of the area.

Pic of the 66%

Z-scores, or standard scores, are a bridge between real data and probabilities surrounding them. We find z-scores with this (important!):

X is the value that we re interested in. We usually want to know the probability of getting a value below or above X.

X is also called the raw score, meaning we haven t prepared it for use at all. Raw as in uncooked.

μ is the mean, in most cases this will be given to you. Look for clues like average, and centered around.

μ is the mean, in most cases this will be given to you. Look for clues like average, and centered around.

σ is the standard deviation, in most cases it s given or computed from SPSS.

The Z-Score is the number of standard deviations above the mean.

Z-Score is also called Standard Score.

Example problem: The time spent on homework in hours/week for full time students is normally distributed with mean 25, and standard deviation = 7 What proportion of students spend more than 20 hours on homework?

Step 1: Identify μ = 25, σ = 7, x = 20. We want the proportion, which is like the probability. We know the distribution is normal. These are clues to find the z-score / standard score, and use it in the z-table to get the proportion.

Step 2: Apply. What do we want?! Z!!!! What do we have?! μ = 25, σ = 7, x = 20.!!!! Use the formula that has Z on one side, and μ, σ, and x on the other.

-0.71 isn t on the table, but by symmetry, we can use 0.71.

By the table, 26.11% is between the mean and z=0.71,23.89% is beyond z=0.71. We want Pr( X > 20), which is Pr(Z > -0.71) Method 1: Split Pr( Z > -0.71) = Pr( Z >0) + Pr(-0.71 < Z < 0) =.5000 +.2611 =.7611 Method 2: Converse Pr( Z > -0.71) = 1 Pr(Z < -0.71) = 1 -.2389 =.7611

We can work backwards from a probability to get a value too, with this: (also important) This is the same formula as the z-score (standard score) formula, but rearranged so that X is the value we get out of it.

Example problem: Homework/week is normally distributed, μ = 25, σ = 7 What s the minimum homework I can expect 90% of the class to do? In other words Pr(X >??? ) =.9000 Step 1: Identify. We have the proportion, and we want the value x. Again, z-score is going to be our bridge.

Going X Z Prob, we used the table last. Going Prob Z X, we ll use the table first. We want the Z value such that 10% of the area is beyond the mean.

As z increases, the area beyond that value decreases. Z % Area Beyond 0.00 50.00 0.01 49.60 0.02 49.20 0.03 48.80 0.04 48.40 0.05 48.01

We can use that to find the Z-score with 10% beyond. (Approximation may be needed) Z % Area Beyond 0.00 50.00 0.01 49.60 0.02 49.20 0.44 33.00 0.45 32.64 0.46 32.36 1.27 10.20 1.28 10.03

Now we know Pr( Z > 1.28) = 10.03%, that s the closest z-score to 10% in the table. What do we want?! X!!!! What do we have?! μ = 25, σ = 7, z = -1.28!!!!

So 90% of the full-time students spend 16.04 hours or more on homework.

What proportion of students spend more than 60 hours/week? μ = 25, σ = 7, x = 60.

Now we have z = 5, how do we get Pr(Z > 5)? The table only goes to z = 3.5ish. Use inference: We want the area beyond z=5, and the area shrinks as z goes up. The smallest area is 0.01%, so the area beyond z=5 must be smaller than that. That s all we can tell from this table. Fewer than 0.01% of students spend 60 hours/week on homework.

(for interest) Very few data points are going to be more than six standard deviations above or below the mean. Far less than 0.01% Six Sigma is a business practice based on making each part in a machine consistent enough that it will work as long as it s within six standard deviations, or 6σ of the mean.

Next time: - A few more notes on Z-scores - Discuss Midterm - We start chapter 6.