Scribd
Upload
33 views

Session 13. Simple Regression

Simple regression analyzes the relationship between a dependent variable and a single independent variable, using a regression equation to determine the strength and nature of the relationship and allow prediction of dependent variable values based on independent variable values. An example shows using regression to analyze the relationship between income and price willing to pay for a car. Key outputs of regression include the R and R-squared values indicating fit, an ANOVA comparing regression and residual variation, and coefficients for the intercept and independent variable.

Uploaded by

Harini Jaish
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
33 views

Session 13. Simple Regression

Simple regression analyzes the relationship between a dependent variable and a single independent variable, using a regression equation to determine the strength and nature of the relationship and allow prediction of dependent variable values based on independent variable values. An example shows using regression to analyze the relationship between income and price willing to pay for a car. Key outputs of regression include the R and R-squared values indicating fit, an ANOVA comparing regression and residual variation, and coefficients for the intercept and independent variable.

Uploaded by

Harini Jaish
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Business Analytics

SIMPLE REGRESSION
Regression
A statistical procedure for analyzing associative relationships between one metric DV and one IV
Used to:
◦ Determine whether a relationship exists
◦ Determine strength of the relationship
◦ Determine mathematical equation relating the IV and DV
◦ Predict the values of the DV
Regression equation

This Photo by Unknown Author is licensed


under CC BY-SA
Regression
Regression equation:

◦ŷi = bo + b1xi
◦ yi denotes observed response for experimental unit i
◦ xi denotes predictor value for experimental unit i
◦ ŷi is the predicted response for experimental unit i
Regression example
Want to determine whether there is a relationship between an individual’s income and the price
they pay for a car
Want to use this information to determine which cars to offer potential customers in new areas
where average income is known
◦ Independent variable = income
◦ Dependent variable = price

ŷi = bo + b1xi
Yi – Price person is willing to pay
Xi – Income of person
Example
H0: (Coefficient of ‘Income of person’) b1 = 0
◦ (meaning Income does not have an association with Price)

H1: b1 ≠ 0
Significance level: 0.05
Regression example
Model summary
Adjusted R Std. error of
Model R R square
square estimate
1 0.873 0.762 0.749 874.779

ANOVA

Model Sum of squares df Mean squares F Sig.


Regression 44182633 1 44182633 57.73 0.000
Residual 1377429 18 765238
Total 57956924 19

Coefficients
Unstandardize Standardized
Model d coefficients coefficients t Sig.

Constant 8286 4.4 0.000

Income 0.56 0.87 7.59 0.00


Regression example
Model summary

Std. error of
Model R R square Adjusted R square
estimate
1
0.873 0.762 0.749 874.779

R – Simple correlation

R square – Variation in DV explained by IV – a value greater than 0.50 is good in engineering

Adjusted R square – Realistic estimation of predictive power of model, based on number of IV in the model –
smaller the delta between R Square and Adjusted R Square the better
Regression example
ANOVA
Model Sum of squares df Mean squares F Sig.
Regression 44182633 1 44182633 57.73 0.000
Residual 1377429 18 765238
Total 57956924 19

F - ratio of Mean S Reg and Mean S Res

Sig. - p-value associated with that F value for the given dF – 2 tailed test – Why?
Regression example
Coefficients

Model Unstandardized Standardized t Sig.


coefficients coefficients
Constant 8286 4.4 0.000
Income 0.56 0.87 7.59 0.00

Constant - Denotes intercept on Y axis

Unstandardized coefficients – for every 1 unit increase in IV, DV will increase by this value

Price = 8286 + 0.56*Income

Standardized coefficients – allows us to compare between IV, when there are multiple IVs.
Exercise 1
Is there a connection between height and weight?

H0: Coefficient = 0
H1: Coefficient ≠ 0
Significance level: 0.05
Exercise 2
Is there is an association between weight of car and mpg?

H0: Coefficient = 0
H1: Coefficient ≠ 0
Significance level: 0.05
Example: Task Cohesion

> model<-
lm(Task_Cohesion_
Cleaned~MC_Formal Formal Task
Coordination Cohesion
Coordination_1,da
ta=mydata)
> summary(model)
Example: Task Cohesion
Call: lm(formula = Task_Cohesion_Cleaned ~ MC_FormalCoordination_1,
data = mydata)

Residuals: Min 1Q Median 3Q Max -2.25651 -0.40387 -0.01057 0.41874


1.42538

Coefficients: Estimate Std. Error t value Pr(>|t|)


(Intercept) 5.41991 0.05982 90.60 <2e-16 ***
MC_FormalCoordination_1 1.02390 0.10180 10.06 <2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6303 on 109 degrees of freedom
Multiple R-squared: 0.4814,
Adjusted R-squared: 0.4766
F-statistic: 101.2 on 1 and 109 DF,
p-value: < 2.2e-16

You might also like