Correlation and Regression

Regression and Correlation
Professor of Biostatistics Department of Epidemiology Director, Data Coordinating Center College of Human Medicine Michigan State University
How do we measure association between two variables?

1. For categorical E and D variables Odds Ratio (OR) Relative Risk (RR) Risk Difference
2. For continuous E & D variables Correlation Coefficient R Coefficient of Determination (R-Square)
Example
A researcher believes that there is a linear relationship between BMI (Kg/m2) of pregnant mothers and the birth-weight (BW in Kg) of their newborn
The following data set provide information on 15 pregnant mothers who were contacted for this study
BMI (Kg/m2) 20 30 50 45 10 30 40 25 50 20 10 55 60 50 35
Birth-weight (Kg) 2.7 2.9 3.4 3.0 2.2 3.1 3.3 2.3 3.5 2.5 1.5 3.8 3.7 3.1 2.8
Scatter Diagram
Scatter diagram is a graphical method to display the relationship between two variables Scatter diagram plots pairs of bivariate observations (x, y) on the X-Y plane
Y is called the dependent variable

X is called an independent variable
Scatter diagram of BMI and Birthweight

4 3.5 3 2.5 2 1.5 1 0.5 0 0 10 20 30 40 50 60 70
Is there a linear relationship between BMI and BW?

Scatter diagrams are important for initial exploration of the relationship between two quantitative variables
In the above example, we may wish to summarize this relationship by a straight line drawn through the scatter of points
Simple Linear Regression

Although we could fit a line "by eye" e.g. using a transparent ruler, this would be a subjective approach and therefore unsatisfactory. An objective, and therefore better, way of determining the position of a straight line is to use the method of least squares. Using this method, we choose a line such that the sum of squares of vertical distances of all points from the line is minimized.
Least-squares or regression line

These vertical distances, i.e., the distance between y values and their corresponding estimated values on the line are called residuals The line which fits the best is called the regression line or, sometimes, the leastsquares line The line always passes through the point defined by the mean of Y and the mean of X
Linear Regression Model

The method of least-squares is available in most of the statistical packages (and also on some calculators) and is usually referred to as linear regression
Y is also known as an outcome variable X is also called as a predictor
Estimated Regression Line
x = 1.775351 + 0.0330187 x + = y . 1.775351 is.called . y int ercept 0.0330187 is.called .the.slope
Application of Regression Line

This equation allows you to estimate BW of other newborns when the BMI is given. e.g., for a mother who has BMI=40, i.e. X = 40 we predict BW to be
x = 1.775351 + 0.0330187 (40) 3.096 + = y
Correlation Coefficient, R
R is a measure of strength of the linear association between two variables, x and y.
Most statistical packages and some hand calculators can calculate R For the data in our Example R=0.94 R has some unique characteristics
Correlation Coefficient, R
R takes values between -1 and +1 R=0 represents no linear relationship between the two variables R>0 implies a direct linear relationship R<0 implies an inverse linear relationship The closer R comes to either +1 or -1, the stronger is the linear relationship
Coefficient of Determination
R2 is another important measure of linear association between x and y (0 R2 1) R2 measures the proportion of the total variation in y which is explained by x For example r2 = 0.8751, indicates that 87.51% of the variation in BW is explained by the independent variable x (BMI).
Difference between Correlation and Regression

Correlation Coefficient, R, measures the strength of bivariate association The regression line is a prediction equation that estimates the values of y for any given x
Limitations of the correlation coefficient

Though R measures how closely the two variables approximate a straight line, it does not validly measures the strength of nonlinear relationship When the sample size, n, is small we also have to be careful with the reliability of the correlation Outliers could have a marked effect on R Causal Linear Relationship
The following data consists of age (in years) and presence or absence of evidence of significant coronary heart disease (CHD) in 100 persons. Code sheet for the data is given as follows:
Serial No. 1. 2. Variable
name
ID AGRP
Variable description
Identification no. Age Group
Codes/values
ID number (unique) 1 = 20-29; 2 = 30-34; 3 = 35-39; 4 = 40-44; 5 = 45-49; 6 = 50-54; 7 = 55-59; 8 = 60-69 in years 0 = Absent; 1 = Present
3. 4.
AGE CHD
Actual age (in years) Presence or absence of CHD
ID
1 2 3 4 5 6 7 8
AGRP
1 1 1 1 1 1 1 1 8 8
AGE
20 23 24 25 25 26 26 28 65 69
CHD
0 0 0 0 1 0 0 0 1 1
99 100
Is there any association between age and CHD?

By categorizing the age variable we will be able to answer the above question the Chi-Square test of independence
Age Group by CHD Age Group Coronary Heart Disease (CHD) Total
Present 40 years >40 years

Total
Absent 32 25
57
7 36
43
39 61
100
Chi-Square Tests Asymp. Sig. (2-sided) 1 1 1 .000 .000 .000 .000 17.434 100 1 .000 .000 Exact Sig. (2-sided) Exact Sig. (1-sided)
Value Pearson Chi-Square Continuitya Correction Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases 17.610 15.919 18.706
b
df
a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 17.16.
Odds Ratio = 0.14 with 95% confidence interval (0.05,0.41) Relative Risk = 0.30 with 95% confidence interval (0.15,0.60)
What about a situation that you do not want to categorize the age?
PLOT OF CHD by AGE
1.2
Presence of Coronary Heart Disease (CHD)
1.0
.8 .6 .4 .2 0.0 -.2 10 20 30 40 50 60 70
Actual age (in years)
Actually, we are interested in knowing whether the probability of having CHD increases by age.
How do you do this?

Frequency Table of Age Group by CHD
Mid point
Age Group of age
CHD n 10 15 12 15 13 08 17 10
100
Mean (proportion) =
Absent 09 13 09 10 07 03 04 02
57
Present 01 02 03 05 06 05 13 08
43
{(Present)/n} (01/10) = 0.10 (02/15) = 0.13 (03/12) = 0.25 (05/15) = 0.33 (06/13) = 0.46 (05/08) = 0.63 (13/17) = 0.76 (08/10) = 0.80
(43/100) = 0.43
20-29 30-34 35-39 40-44 45-49 50-54 55-59 60-69 Total
25 32.5 37.5 42.5 47.5 52.5 57.5 65
Logistic Regression
Logistic Regression is used when the outcome variable is categorical The independent variables could be either categorical or continuous The slope coefficient in the Logistic Regression Model has a relationship with the OR Multiple Logistic Regression model can be used to adjust for the effect of other variables when assessing the association between E & D variables

Correlation and Regression

Uploaded by

Copyright:

Available Formats

Correlation and Regression

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Correlation and Regression

Uploaded by

Copyright:

Available Formats

Regression and Correlation

How do we measure association between two variables?

Y is called the dependent variable

Scatter diagram of BMI and Birthweight

Is there a linear relationship between BMI and BW?

Simple Linear Regression

Least-squares or regression line

Linear Regression Model

Estimated Regression Line

x = 1.775351 + 0.0330187 x + = y . 1.775351 is.called . y int ercept 0.0330187 is.called .the.slope

Application of Regression Line

x = 1.775351 + 0.0330187 (40) 3.096 + = y

Difference between Correlation and Regression

Limitations of the correlation coefficient

Actual age (in years) Presence or absence of CHD

Is there any association between age and CHD?

Present 40 years >40 years

Presence of Coronary Heart Disease (CHD)

Actual age (in years)

How do you do this?

20-29 30-34 35-39 40-44 45-49 50-54 55-59 60-69 Total

25 32.5 37.5 42.5 47.5 52.5 57.5 65

You might also like