Download presentation
Presentation is loading. Please wait.
Published byDamon Greer Modified over 9 years ago
1
Last Time Interpretation of Confidence Intervals Handling unknown μ and σ T Distribution Compute with TDIST & TINV (Recall different organization) (relative to NORMDIST & NORMINV)
2
Reading In Textbook Approximate Reading for Today’s Material: Pages 420-427, 86-94 Approximate Reading for Next Class: Pages 101-105, 447-465, 511-516
3
Deeper look at Inference Recall: “inference” = CIs and Hypo Tests Main Issue: In sampling distribution Usually σ is unknown, so replace with an estimate, s. For n large, should be “OK”, but what about: n small? How large is n “large”?
4
Unknown SD Then So can write: Replace by, then has a distribution named: “t-distribution with n-1 degrees of freedom”
5
t - Distribution Notes: 4.Calculate t probs (e.g. areas & cutoffs), using TDIST & TINV Caution: these are set up differently from NORMDIST & NORMINV
6
EXCEL Functions Summary: Normal: plug in: get out: NORMDIST: cutoff area NORMINV: area cutoff (but TDIST is set up really differently)
7
EXCEL Functions t distribution: Area 2 tail: plug in: get out: TDIST: cutoff area TINV: area cutoff (EXCEL note: this one has the inverse)
8
t - Distribution Application 1: Confidence Intervals
9
t - Distribution Application 1: Confidence Intervals Recall:
10
t - Distribution Application 1: Confidence Intervals Recall: margin of error
11
t - Distribution Application 1: Confidence Intervals Recall: margin of error from NORMINV
12
t - Distribution Application 1: Confidence Intervals Recall: margin of error from NORMINV or CONFIDENCE
13
t - Distribution Application 1: Confidence Intervals Recall: margin of error from NORMINV or CONFIDENCE Using TINV?
14
t - Distribution Application 1: Confidence Intervals Recall: margin of error from NORMINV or CONFIDENCE Using TINV? Careful need to standardize
15
t - Distribution Using TINV? Careful need to standardize
16
t - Distribution Using TINV? Careful need to standardize
17
t - Distribution Using TINV? Careful need to standardize
18
t - Distribution Using TINV? Careful need to standardize # spaces on number line
19
t - Distribution Using TINV? Careful need to standardize # spaces on number line Need to work in to use TINV
20
t - Distribution Using TINV? Careful need to standardize # spaces on number line Need to work in to use TINV
21
t - Distribution
22
distribution
23
t - Distribution distribution
24
t - Distribution distribution So want:
25
t - Distribution distribution So want: i.e. want:
26
t - Distribution Class Example 15, Part I http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg15.xls Old text book problem 7.24
27
t - Distribution Class Example 15, Part I http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg15.xls Old text book problem 7.24: In a study of DDT poisoning, researchers fed several rats a measured amount. They measured the “absolutely refractory period” required for a nerve to recover after a stimulus. Measurements on 4 rats gave:
28
t - Distribution Class Example 15, Part I http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg15.xls Old text book problem 7.24: Measurements on 4 rats gave: 1.6 1.7 1.8 1.9 a)Find the mean refractory period, and the standard error of the mean b)Give a 95% CI for the mean “absolutely refractory period” for all rats of this strain
29
t - Distribution Class Example 15, Part I Data in cells B9:E9
30
t - Distribution Class Example 15, Part I Data in cells B9:E9 Note: small sample size (n = 4)
31
t - Distribution Class Example 15, Part I Data in cells B9:E9 Note: small sample size (n = 4), population sd, σ, unknown
32
t - Distribution Class Example 15, Part I Data in cells B9:E9 Note: small sample size (n = 4), population sd, σ, unknown, so use sample sd, s
33
t - Distribution Class Example 15, Part I Data in cells B9:E9 Note: small sample size (n = 4), population sd, σ, unknown, so use sample sd, s, and t distribution
34
t - Distribution Class Example 15, Part I Data in cells B9:E9 Center CI at Sample Mean
35
t - Distribution Class Example 15, Part I Data in cells B9:E9 Center CI at Sample Mean Measure Sample Spread by S. D.
36
t - Distribution Class Example 15, Part I Data in cells B9:E9 Center CI at Sample Mean Measure Sample Spread by S. D. Divide by to get Standard Error
37
t - Distribution Class Example 15, Part I Data in cells B9:E9 Center CI at Sample Mean Measure Sample Spread by S. D. Divide by to get Standard Error Which answers (a)
38
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9
39
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error
40
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compute using TINV
41
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Recall: d.f. = n – 1 = 4 – 1
42
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compare to old Normal CIs
43
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compare to old Normal CIs Compute using CONFIDENCE
44
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compare to old Normal CIs T CIs are wider
45
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compare to old Normal CIs T CIs are wider (as expected)
46
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compare to old Normal CIs Left End
47
t - Distribution Class Example 15, Part I (b) 95% CI for μ Data in cells B9:E9 CI Radius = Margin of Error Compare to old Normal CIs Left End Right End
48
t - Distribution Confidence Interval HW: 7.24 (a. Q-Q roughly linear, so OK, b. 43.17, 4.41, 0.987 c. [41.1, 45.2]) 7.25
49
And now for something completely different An extreme “sport” video:
50
t - Distribution Application 2: Hypothesis Tests
51
t - Distribution Application 2: Hypothesis Tests Idea: Calculate P-values using TDIST
52
t - Distribution Application 2: Hypothesis Tests Idea: Calculate P-values using TDIST
53
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example
54
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example Recall Data in cells B9:E9
55
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example Recall Data in cells B9:E9 As above: t – distribution appropriate
56
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example Recall Data in cells B9:E9 As above: t – distribution appropriate (small sample, and using s ≈ σ)
57
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example, Suppose that the mean “absolutely refractory period” is known to be 1.3
58
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example, Suppose that the mean “absolutely refractory period” is known to be 1.3 (recall observed in data)
59
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example, Suppose that the mean “absolutely refractory period” is known to be 1.3. DDT poisoning should slow nerve recovery, and so increase this period.
60
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 For the above DDT poisoning example, Suppose that the mean “absolutely refractory period” is known to be 1.3. DDT poisoning should slow nerve recovery, and so increase this period. Do the data give good evidence for this supposition?
61
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 Let = population mean absolutely refractory period for poisoned rats.
62
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 Let = population mean absolutely refractory period for poisoned rats. (checking strong evidence for this)
63
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 Let = population mean absolutely refractory period for poisoned rats. (from before)
64
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry}
65
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry}
66
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry}
67
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry}
68
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry}
69
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry} Now use TDIST
70
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry} Now use TDIST
71
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry} Now use TDIST
72
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry} Now use TDIST
73
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry} Now use TDIST Degrees of Freedom = n – 1 = 4 - 1
74
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 P-value = P{what saw or more conclusive | H 0 – H A Bdry} Now use TDIST Tails
75
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 From Class Example 27, part 2: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg15.xls P - value = 0.003
76
t – Distribution Hypo Testing E.g. Old Textbook Example 7.26 From Class Example 27, part 2: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg15.xls P - value = 0.003 Interpretation: very strong evidence, for either yes-no or gray-level
77
t – Distribution Hypo Testing Variations: For “opposite direction” hypotheses:
78
t – Distribution Hypo Testing Variations: For “opposite direction” hypotheses: P-value =
79
t – Distribution Hypo Testing Variations: For “opposite direction” hypotheses: P-value =
80
t – Distribution Hypo Testing Variations: For “opposite direction” hypotheses: P-value = [wrong way for TDIST(…,1)]
81
t – Distribution Hypo Testing Variations: For “opposite direction” hypotheses: P-value = Then use symmetry
82
t – Distribution Hypo Testing Variations: For “opposite direction” hypotheses: P-value = Then use symmetry, i.e. put - into TDIST.
83
t – Distribution Hypo Testing Variations: For 2-sided hypotheses
84
t – Distribution Hypo Testing Variations: For 2-sided hypotheses: H 0 : μ = H 1 : μ ≠
85
t – Distribution Hypo Testing Variations: For 2-sided hypotheses: H 0 : μ = H 1 : μ ≠
86
t – Distribution Hypo Testing Variations: For 2-sided hypotheses: H 0 : μ = H 1 : μ ≠ Use 2-tailed version of TDIST
87
t – Distribution Hypo Testing Variations: For 2-sided hypotheses: H 0 : μ = H 1 : μ ≠ Use 2-tailed version of TDIST, i.e. TDIST(…,2)
88
t – Distribution Hypo Testing HW: Interpret P-values: (i)yes-no (ii)gray-level 7.21e ((i)significant, (ii) significant, but not very strongly so) 7.22e (0.0619, (i) not significant (ii) not sig., but nearly significant)
89
Research Corner Another SiZer analysis: Mollusk Extinction Data
90
Research Corner Another SiZer analysis: Mollusk Extinction Data From Matthew Campbell UNC Master’s Student In Geological Sciences
91
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells
92
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells (fossil beds up and down Eastern Seaboard)
93
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells Dated (by fossil bed)
94
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells Dated (by fossil bed) Biologically categorized
95
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells Dated (by fossil bed) Biologically categorized (family – genus – species, etc.)
96
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells Dated (by fossil bed) Biologically categorized Goal: study extinctions over long periods
97
Research Corner Another SiZer analysis: Mollusk Extinction Data Data points: Fossilized shells Dated (by fossil bed) Biologically categorized Goal: study extinctions over long periods (via last time saw each)
98
Research Corner Another SiZer analysis: Mollusk Extinction Data Oversmoothed: nothing interesting
99
Research Corner Another SiZer analysis: Mollusk Extinction Data Undersmoothed: many bumps appear
100
Research Corner Another SiZer analysis: Mollusk Extinction Data Undersmoothed: many bumps appear but not statistically significant
101
Research Corner Another SiZer analysis: Mollusk Extinction Data Intermediate Smoothing two bumps appear
102
Research Corner Another SiZer analysis: Mollusk Extinction Data Intermediate Smoothing two bumps appear SiZer result:not statistically significant
103
Research Corner Another SiZer analysis: Mollusk Extinction Data Matthew’s Comment: Whoah, those are times of mass extinctions
104
Research Corner Another SiZer analysis: Mollusk Extinction Data Matthew’s Comment: Whoah, those are times of mass extinctions Any way to show these are “really there”?
105
Research Corner Another SiZer analysis: Mollusk Extinction Data Any way to show these are “really there”?
106
Research Corner Another SiZer analysis: Mollusk Extinction Data Any way to show these are “really there”? Standard Answer: Get more data
107
Research Corner Another SiZer analysis: Mollusk Extinction Data Any way to show these are “really there”? Challenge: Took 100s of year to get these!
108
Research Corner Another SiZer analysis: Mollusk Extinction Data Any way to show these are “really there”? Alternate Approach: Refined from Genus level to Species
109
Research Corner Another SiZer analysis: Mollusk Extinction Data Species level result
110
Research Corner Another SiZer analysis: Mollusk Extinction Data Species level result: Now both bumps are significant
111
Research Corner Another SiZer analysis: Mollusk Extinction Data Species level result: Now both bumps are significant Consistent with Global Climactic Events
112
Variable Relationships Chapter 2 in Text
113
Variable Relationships Chapter 2 in Text Idea: Look beyond single quantities
114
Variable Relationships Chapter 2 in Text Idea: Look beyond single quantities, to how quantities relate to each other.
115
Variable Relationships Chapter 2 in Text Idea: Look beyond single quantities, to how quantities relate to each other. E.g. How do HW scores “relate” to Exam scores?
116
Variable Relationships Chapter 2 in Text Idea: Look beyond single quantities, to how quantities relate to each other. E.g. How do HW scores “relate” to Exam scores? Section 2.1: Useful graphical device: Scatterplot
117
Plotting Bivariate Data Toy Example:Ordered pairs (1,2) (3,1) (-1,0) (2,-1)
118
Plotting Bivariate Data Toy Example:Ordered pairs Captures relationship between X & Y (1,2)as (X,Y) (3,1) (-1,0) (2,-1)
119
Plotting Bivariate Data Toy Example:Ordered pairs Captures relationship between X & Y (1,2)as (X,Y) (3,1)e.g. (height, weight) (-1,0) (2,-1)
120
Plotting Bivariate Data Toy Example:Ordered pairs Captures relationship between X & Y (1,2)as (X,Y) (3,1)e.g. (height, weight) (-1,0)e.g. (MT Score, Final Exam Score) (2,-1)
121
Plotting Bivariate Data Toy Example: (1,2)Think in terms of: (3,1) (-1,0) X coordinates (2,-1)
122
Plotting Bivariate Data Toy Example: (1,2)Think in terms of: (3,1) (-1,0) X coordinates (2,-1)Y coordinates
123
Plotting Bivariate Data Toy Example: (1,2)Think in terms of: (3,1) (-1,0) X coordinates (2,-1)Y coordinates And plot in x,y plane, to see relationship
124
Plotting Bivariate Data Toy Example: (1,2) (3,1) (-1,0) (2,-1)
125
Plotting Bivariate Data Toy Example: (1,2) (3,1) (-1,0) (2,-1)
126
Plotting Bivariate Data Toy Example: (1,2) (3,1) (-1,0) (2,-1)
127
Plotting Bivariate Data Toy Example: (1,2) (3,1) (-1,0) (2,-1)
128
Plotting Bivariate Data Sometimes: Can see more insightful patterns by connecting points
129
Plotting Bivariate Data Sometimes: Useful to switch off points, and only look at lines/curves
130
Plotting Bivariate Data Common Name: “Scatterplot”
131
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel
132
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab
133
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab Charts
134
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab Charts Scatter Button
135
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab Charts Scatter Button Choose Dots
136
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab Charts Scatter Button Choose Dots (but note other options)
137
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab Charts Scatter Button Choose Dots Manipulate plot as done before for bar plots
138
Plotting Bivariate Data Common Name: “Scatterplot” A look under the hood in Excel: Insert Tab Charts Scatter Button Choose Dots Manipulate plot as done before for bar plots (e.g. titles, labels, colors, styles, …)
139
Scatterplot E.g. Class Example 16: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Data from related Intro. Statistics Class
140
Scatterplot E.g. Class Example 16: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Data from related Intro. Statistics Class (actual scores)
141
Scatterplot E.g. Class Example 16: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Data from related Intro. Statistics Class (actual scores) A.How does HW score predict Final Exam?
142
Scatterplot E.g. Class Example 16: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Data from related Intro. Statistics Class (actual scores) A.How does HW score predict Final Exam? x i = HW, y i = Final Exam
143
Scatterplot E.g. Class Example 16: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Data from related Intro. Statistics Class (actual scores) A.How does HW score predict Final Exam? x i = HW, y i = Final Exam (Study Relationship using scatterplot)
144
Scatterplot E.g. Class Example 16: How does HW score predict Final Exam? x i = HW, y i = Final Exam
145
Scatterplot E.g. Class Example 16: How does HW score predict Final Exam? x i = HW, y i = Final Exam (Scatterplot View)
146
Scatterplot E.g. Class Example 16: How does HW score predict Final Exam? x i = HW, y i = Final Exam i.In top half of HW scores
147
Scatterplot E.g. Class Example 16: How does HW score predict Final Exam? x i = HW, y i = Final Exam i.In top half of HW scores: Better HW Better Final
148
Scatterplot E.g. Class Example 16: How does HW score predict Final Exam? x i = HW, y i = Final Exam i.In top half of HW scores: Better HW Better Final ii.For lower HW
149
Scatterplot E.g. Class Example 16: How does HW score predict Final Exam? x i = HW, y i = Final Exam i.In top half of HW scores: Better HW Better Final ii.For lower HW: Final is more “random”
150
Scatterplots Common Terminology: When thinking about “X causes Y”,
151
Scatterplots Common Terminology: When thinking about “X causes Y”, Call X the “Explanatory Var.”
152
Scatterplots Common Terminology: When thinking about “X causes Y”, Call X the “Explanatory Var.” or “Indep. Var.”
153
Scatterplots Common Terminology: When thinking about “X causes Y”, Call X the “Explanatory Var.” or “Indep. Var.” Call Y the “Response Var.”
154
Scatterplots Common Terminology: When thinking about “X causes Y”, Call X the “Explanatory Var.” or “Indep. Var.” Call Y the “Response Var.” or “Dep. Var.”
155
Scatterplots Common Terminology: When thinking about “X causes Y”, Call X the “Explanatory Var.” or “Indep. Var.” Call Y the “Response Var.” or “Dep. Var.” (think of “Y as function of X”)
156
Scatterplots Common Terminology: When thinking about “X causes Y”, Call X the “Explanatory Var.” or “Indep. Var.” Call Y the “Response Var.” or “Dep. Var.” (think of “Y as function of X”) (although not always sensible)
157
Scatterplots Note: Sometimes think about causation
158
Scatterplots Note: Sometimes think about causation, Other times: “Explore Relationship”
159
Scatterplots Note: Sometimes think about causation, Other times: “Explore Relationship” HW: 2.9
160
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1
161
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 (Replace Final above with 1 st Midterm)
162
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1
163
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 i.Better HW better Exam (general upwards tendency still the same)
164
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 i.Better HW better Exam ii.Wider range MT1 scores (for each range of HW scores) (relative to final scores)
165
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 i.Better HW better Exam ii.Wider range MT1 scores iii.HW doesn’t predict MT1 (as well as HW predicted the final)
166
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 i.Better HW better Exam ii.Wider range MT1 scores iii.HW doesn’t predict MT1 iv.“Outliers” in scatterplot
167
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 i.Better HW better Exam ii.Wider range MT1 scores iii.HW doesn’t predict MT1 iv.“Outliers” in scatterplot e.g. HW = 72, MT1 = 94
168
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls B.How does HW predict Midterm 1? x i = HW, y i = MT1 i.Better HW better Exam ii.Wider range MT1 scores iii.HW doesn’t predict MT1 iv.“Outliers” in scatterplot may not be outliers in either individual variable e.g. HW = 72, MT1 = 94 (bad HW, but good MT1?, fluke???)
169
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2
170
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 (Different choice of x and y, since studying different relationship)
171
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 (Study Relationship using tool of scatterplot)
172
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” (don’t expect better MT1 to lead to better MT2)
173
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 (again clear overall upwards trend)
174
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 iii.Wider range of MT2 (for each range of MT1)
175
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 iii.Wider range of MT2 i.e. “not good predictor” (MT1)(of MT2)
176
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 iii.Wider range of MT2 i.e. “not good predictor” iv.Interesting Outliers
177
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 iii.Wider range of MT2 i.e. “not good predictor” iv.Interesting Outliers: MT1 = 100, MT2 = 56 (oops!)
178
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 iii.Wider range of MT2 i.e. “not good predictor” iv.Interesting Outliers: MT1 = 100, MT2 = 56 MT1 = 23, MT2 = 74 (woke up!)
179
Class Scores Scatterplots http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls C.How does MT1 predict MT2? x i = MT1, y i = MT2 i.Idea: less “causation”, more “exploration” ii.High MT1 High MT2 iii.Wider range of MT2 i.e. “not good predictor” iv.Interesting Outliers: MT1 = 100, MT2 = 56 MT1 = 23, MT2 = 74 MT1 70s, MT2 90s (moved up!)
180
And now for something completely different A thought provoking movie clip: http://www.aclu.org/pizza/
181
Important Aspects of Relations I.Form of Relationship (Linear or not?)
182
Important Aspects of Relations I.Form of Relationship II.Direction of Relationship (trending up or down?)
183
Important Aspects of Relations I.Form of Relationship II.Direction of Relationship III.Strength of Relationship (how much of data “explained”?)
184
I.Form of Relationship Linear: Data approximately follow a line
185
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls
186
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls
187
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final vs. High values of HW is “best”
188
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final vs. High values of HW is “best” But saw others with “rough linear trend”
189
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final vs. High values of HW is “best” But saw others with “rough linear trend”
190
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final vs. High values of HW is “best” But saw others with “rough linear trend” Interesting question: Measure strength of linear trend
191
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final vs. High values of HW is “best” Nonlinear: Data follows different pattern (non-linear)
192
I.Form of Relationship Linear: Data approximately follow a line Previous Class Scores Example http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final vs. High values of HW is “best” Nonlinear: Data follows different pattern Nice Example: Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls
193
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci.
194
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci.
195
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate
196
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago
197
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago: Small shells from ocean floor cores
198
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago: Small shells from ocean floor cores Ratios of Isotopes of Strontium
199
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago: Small shells from ocean floor cores Ratios of Isotopes of Strontium Reflects Ice Ages, via Sea Level
200
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago: Small shells from ocean floor cores Ratios of Isotopes of Strontium Reflects Ice Ages, via Sea Level (50 meter difference!)
201
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago: Small shells from ocean floor cores Ratios of Isotopes of Strontium Reflects Ice Ages, via Sea Level (50 meter difference!) As function of time
202
Bralower’s Fossil Data http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg17.xls From T. Bralower, formerly of Geological Sci.T. BralowerGeological Sci. Studies Global Climate, millions of years ago: Small shells from ocean floor cores Ratios of Isotopes of Strontium Reflects Ice Ages, via Sea Level (50 meter difference!) As function of time Clearly nonlinear relationship
203
II. Direction of Relationship Positive Association
204
II. Direction of Relationship Positive Association X bigger Y bigger
205
II. Direction of Relationship Positive Association X bigger Y bigger E.g. Class Scores Data above
206
II. Direction of Relationship Positive Association X bigger Y bigger Negative Association
207
II. Direction of Relationship Positive Association X bigger Y bigger Negative Association X bigger Y smaller
208
II. Direction of Relationship Positive Association X bigger Y bigger Negative Association X bigger Y smaller E.g. X = alcohol consumption, Y = Driving Ability Clear negative association
209
II. Direction of Relationship Positive Association X bigger Y bigger Negative Association X bigger Y smaller Note: Concept doesn’t always apply:
210
II. Direction of Relationship Positive Association X bigger Y bigger Negative Association X bigger Y smaller Note: Concept doesn’t always apply: Bralower’s Fossil Data
211
III. Strength of Relationship Idea: How close are points to lying on a line? Revisit Class Scores Example: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls
212
III. Strength of Relationship Idea: How close are points to lying on a line? Revisit Class Scores Example: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final Exam is “closely related to HW”
213
III. Strength of Relationship Idea: How close are points to lying on a line? Revisit Class Scores Example: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final Exam is “closely related to HW” Midterm 1 less closely related to HW
214
III. Strength of Relationship Idea: How close are points to lying on a line? Revisit Class Scores Example: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final Exam is “closely related to HW” Midterm 1 less closely related to HW Midterm 2 even less related to Midterm 1
215
III. Strength of Relationship Idea: How close are points to lying on a line? Revisit Class Scores Example: http://www.stat-or.unc.edu/webspace/courses/marron/UNCstor155-2009/ClassNotes/Stor155Eg16.xls Final Exam is “closely related to HW” Midterm 1 less closely related to HW Midterm 2 even less related to Midterm 1 Interesting Issue: Measure this strength
216
Linear Relationship HW HW: 2.11, 2.13, 2.15, 2.17
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.