Scribd
Upload
77 views3 pages

Chapter 5 - Correlation and Regression

The document discusses correlation, linear regression, and how to interpret regression analysis results. It defines correlation, scatter plots, and the product moment correlation coefficient (PMCC) and how they relate to the strength of correlation between two variables. It also defines independent and dependent variables, the requirements and limitations of linear regression, and how to find and interpret the linear regression line and use it to predict values.

Uploaded by

Sayar Hein
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
77 views3 pages

Chapter 5 - Correlation and Regression

The document discusses correlation, linear regression, and how to interpret regression analysis results. It defines correlation, scatter plots, and the product moment correlation coefficient (PMCC) and how they relate to the strength of correlation between two variables. It also defines independent and dependent variables, the requirements and limitations of linear regression, and how to find and interpret the linear regression line and use it to predict values.

Uploaded by

Sayar Hein
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Chapter 5 - Correlation and Regression

Correlation
Scatter Diagrams

• Correlation is strong if the points mostly lie on a straight line.


• Correlation is weak if the points are spread out away from the regression line.

Product Moment Correlation Coefficient (PMCC)


• Denoted by r
• If | r | is close to 1, the correlation is strong
• If | r | is close to 0, the correlation is weak or there is no correlation
• Coding has no effect on PMCC

Sx y
r = , |r | ⩽ 1
Sx x Syy

∑ x∑ y
∑ ∑
Sx y = (x − x̄ )(y − ȳ) = xy −
n
2
( ∑ x)
(x − x̄ )2 = x2 −
∑ ∑
Sx y =
n

Answering Correlation Written Questions


• Find out correlation either from the scatter diagram or PMCC
• Interpret the results of the question in addition to answering whether the correlation is positive or
negative.
Independent and Dependent Variables
Independent Variables
• Also called explanatory variables
• Can be set independently of the other variable.
• Plotted along the x-axis

Dependent Variables
• Also called response variables
• Values determined by the values of the independent variable
• Plotted along the y-axis

Causal Relationship
• Two variables are in a causal relationship if a change in one variable causes a change in the other
variable.
• Use common sense to discern whether two variables are in a causal relationship.

Linear Regression
Requirements for using linear regression
• Strong correlation ( | r | close to 1)
• Linear correlation (Data resembles a straight line in the scatter plot)

When is Linear Regression Unreliable?


• If we have a linear regression line for predicting y given x, we cannot use it to predict x given y.
• Linear regression line is unreliable if the x to use is outside the range of given dataset (extrapolation)
• Linear regression line is reliable if the x to use is inside the range of given dataset (interpolation)

Least Square Regression Line


y = a + bx
Sx y
b=
Sx x

a = ȳ + b x̄
Interpreting y = a + bx
a is the value of y when x is zero.
b is the amount by which y increases for an increase of 1 in x.

You must write your interpretation in context − copy out the question word for word.
Coding and Regression Lines
Given: a coded to x
b coded to y
dataset for x and y

Question: Predict b given a

Solution:
1. Find regression line for x and y
2. Convert a into x
3. Find the value of y with the regression line and the value of x
4. Convert y into b

You might also like