Scribd
Upload
98 views

Extra Notes On Regression

The document discusses using a scientific calculator to perform linear regression analysis on sample data sets. It provides steps for entering paired (x, y) data into a CASIO fx-570MS calculator, describes the calculator operators used to obtain regression outputs like the sum of x-squared, regression coefficients alpha (intercept) and beta (slope), correlation coefficient r, and mean x-bar and y-bar. An example calculates the linear regression equation from a provided data set. Two tasks ask the reader to find the linear regression model and correlation coefficient for a second data set, and predict the y-value for a given x-value in that second data set.

Uploaded by

Luciano Ebang
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
98 views

Extra Notes On Regression

The document discusses using a scientific calculator to perform linear regression analysis on sample data sets. It provides steps for entering paired (x, y) data into a CASIO fx-570MS calculator, describes the calculator operators used to obtain regression outputs like the sum of x-squared, regression coefficients alpha (intercept) and beta (slope), correlation coefficient r, and mean x-bar and y-bar. An example calculates the linear regression equation from a provided data set. Two tasks ask the reader to find the linear regression model and correlation coefficient for a second data set, and predict the y-value for a given x-value in that second data set.

Uploaded by

Luciano Ebang
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Correlation and Linear Regression

USING A SCIENTIFIC CALCULATOR

Most scientific calculators provide tools for obtaining the estimated regression coefficients
and hence the fitted regression line. Calculator type : CASIO fx-570MS.

1. Choose the Regression mode :

Mode Mode 2

2. Choose Linear regression :

3. Clear the memory :

Shift CLR 1 =

4. Enter the pair of ( xi , yi ) values :

xi , yi M+

5. Continue Step 4 for all pairs of ( xi , yi ) , i = 1 , 2, , 3 , 4 , , n.

NOTE : Step 3 is vitally important when storing a new data set so that the old data set will
be removed and will not be mixed with the new data set to ensure an accurate analysis.

Chapter : Correlation and Linear Regression

1
Correlation and Linear Regression

Once the sample data set are stored in the calculator, we can retrieve the available output
by pressing the appropriate operators as shown in TABLE 2 below :

Operators Output

Shift S-SUM 1 = x2
Shift S-SUM 2 = x

Shift S-SUM 3 = n

Shift S-SUM 1 = y2
Shift S-SUM 2 = y

Shift S-SUM 3 = xy

Shift S-VAR 1 = x

Shift S-VAR 1 = y

Shift S-VAR 1 =

Shift S-VAR 2 =

Shift S-VAR 3 = r Chapter : Correlation and Linear Regression

TABLE 2

NOTE : is the intercept value of the regression line and is the slope of the regression
line.

2
Correlation and Linear Regression

EXAMPLE 1

Obtain the equation of the linear regression y on x for the following data :

x 25 30 35 40 45 50
y 79 71 66 56 47 43

To obtain the regression line, follow these steps :

1. Follow the five steps in page 1.

2. At Step 3, before we store the new data set, we must always make sure that
the old data set is already cleared. This is indicated by n 0 on the calculator
screen before the new data set is stored.

3. After storing the above data set, we should obtain the following output :

Operators Output

Shift S-SUM 1 = x2 8875

Shift S-SUM 2 = x 225

Shift S-SUM 3 = n6

Shift S-SUM 1 = y2 22832

Shift S-SUM 2 = y 362

Shift S-SUM 3 = xy 12920

x 37.5
Chapter : Correlation and Linear Regression

Shift S-VAR 1 =

Shift S-VAR 1 = y 60.3333

Shift S-VAR 1 = 116.4762

Shift S-VAR 2 = 1.4971

The regression line is


y 116.4762 1.4971x

3
Correlation and Linear Regression

TASK 1

An article in the Journal of Sound and Vibration (Vol. 151, 1991, pp. 383 394) described a
study which investigated the relationship between noise exposure and hypertension. The
nois exposure is measured by the sound pressure level (SPL) in decibels, whereas
hyperteension is measured by the blood pressure rise (BPR) in millileters of mercury
(mmHg). A representative data set reported by Montgomery & Runger (2003), p 382, is as
follows :
SPL, x 60 63 65 70 70 70 80 90 80 80
BPR, y 1 0 1 2 5 1 4 6 2 3

SPL, x 85 89 90 90 90 90 94 100 100 100


BPR, y 5 4 6 8 4 5 7 9 7 6

Obtain the linear regression model of y on x.

Answer : y 10.1315 0.1743 x

TASK 2

Refer to EXAMPLE 1. Obtain the correlation coefficient to measure the relationship between
the variables. Comment your answer.

TASK 3

Refer to TASK 1. Predict the blood pressure rise when the sound pressure level is 98
decibels.

Chapter : Correlation and Linear Regression

You might also like