Please enable JavaScript.
Coggle requires JavaScript to display documents.
Correlation & Regression (limited), All Rights Reserved by Brent…
Correlation & Regression (limited)
Correlation
Measures strength of Association
does not imply causation
Viz Tool (click here)
Simple Regression
Prediction w/ 1 explanatory variable
Regression/Prediction line
How chosen
Error Term & SSE
Interpreting slope (b1)
Predicting (word problem)
Test if Explanatory variable is significant
R-Squared (How good of a fit?)
Formula / Interpretation
Example
Example with Data
Health costs Dataset for Tableau (click here)
HW due 2/28/21
Multiple Regression
2 or more explanatory variables
Example with Data
Health costs Dataset for Tableau (click here)
Regressing 'Charges' on 'BMI' & 'Age'
Regressing 'Charges' on 'BMI' & 'Children'
Regressing 'Charges' on 'Age' & 'Smoker'
'Smoker' is a
Dummy Variable
Dummy Variable is
nominal
qualitative
Qualitative is nominal or ordinal
Pivoting ANOVA to a Dummy Variable
Example 1
Example 2
Dummy Variable effect
HW due 3/2/21
Visualizing Regression of 3 explanatory variables
Health costs Dataset for Excel (click here)
Multiple Regression with 6 explanatory variables
Three QuantitativeVariables
Three Dummy Variables
How to interpret Dummy effects
Using R-Studio
R-squared & Slope interpretation
Which are significant?
Comparing Models
Using
R-squared
: compare with same # of variables
Using
Adjusted R-squared
: compare when different # of variables
HW due 3/7/21
Chow Test
("Structural Break" in data)
2 separate regression lines instead of 1
Example
Question?
Visual of 1 Regression line (Regressing HPI on Years)
Hypothesis
F-test statistic
Conclusion
Visual of 2 Regression lines
1 more item...
Examples
When to use
All Rights Reserved by Brent Marinan & University of Arizona
1.
2.
3.
4.
a.
b.
c.
a.
b.
c.
a.
b.
c.
b.
c.
i.
ii.
iii.
iv.
i.
ii.
a.
00:00 (2/24/21)
01:42 (2/24/21)
05:12 (2/24/21)
15:03 (2/24/21)
17:24 (2/24/21)
18:29 (2/24/21)
21:20 (2/24/21)
23:00 (2/24/21)
26:45 (2/24/21)
29:35 (2/24/21)
35:55 (2/24/21)
38:57 (2/24/21)
41:38 (2/24/21)
Effect of outliers
1.
2.
3.
4.
5.
1.
2.
3.
00:01 (3/1/21)
01:26 (3/1/21)
06:38 (3/1/21)
09:36 (3/1/21)
14:20 (3/1/21)
17:14 (3/1/21)
18:34 (3/1/21)
19:29 (3/1/21)
20:25 (3/1/21)
22:31 (3/1/21)
24:34 (3/1/21)
24:32 (3/1/21)
29:18 (3/1/21)
31:05 (3/1/21)
00:01 (3/3/21)
19:48 (3/3/21)
22:56 (3/3/21)
28:26 (3/3/21)
00:01 (3/8/21)
01:50 (3/8/21)
04:50 (3/8/21)
06:10 (3/8/21)
07:18 (3/8/21)
08:10 (3/8/21)
