0% found this document useful (0 votes)
3 views7 pages

Simple Linear Regression Exercises

This document contains 21 exercises on basic concepts of simple linear regression such as the slope and the correlation coefficient of the regression line, the explained variance, and how to use linear regression to predict values and analyze the relationship between variables. The exercises ask to calculate regression equations, percentages of explained variance, and predicted values for different datasets.
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
0% found this document useful (0 votes)
3 views7 pages

Simple Linear Regression Exercises

This document contains 21 exercises on basic concepts of simple linear regression such as the slope and the correlation coefficient of the regression line, the explained variance, and how to use linear regression to predict values and analyze the relationship between variables. The exercises ask to calculate regression equations, percentages of explained variance, and predicted values for different datasets.
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

Simple linear regression 109

EXERCISES
1. Why are the signs of the correlation coefficient and the slope equal?
of the least squares regression line? Rp.'r=bsx/sr.

2. Given the least squares regression lines Y=a+bX y


X = c + dY. Verify that b is equal to the coefficient of determination.

3. If the graphs of the regression lines of Y on X and of X on Y form an angle


At 90 degrees, what can be said about the correlation index? Rp. it is zero.

4. If (JCj JC2•>'2)»---»3Vi)arepairsofobserveddatathatare
found on the line L:Y=mX+b, why L is the regression line
least squares for these points?, what percentage of the total variance of
the>',•isexplainedbyL?.
Rp. why E (y ->, ) 2=0, 100%.

5. Given the least squares regression line Y=a+bX, if a is produced a


Increase equals one of the values of X, how much is the increase?
respective that is produced in K? [Link].

6 AI perform the YennX regression based on a sample of 10 pairs of data.


It is known that the variance of the it is equal to 16 and that the sum of
squares due to the regression 140. What percentage of the variance of the y,
Is it explained by regression? Rp. 87.5%.

The correlation coefficient between two variables X and Y is r = 0.60. If x = 1.50,


5 }/ = 2 . 0 0 ,x = 1 0 ,y= 2 0 find the regression line:
a)ofYenX,b)ofXenY.
K = 12 + 0.8X

8 The pairs (x1, y1), ..., (xn, yn) are such that they satisfy the relationship

Ify=0is,therelationshipSXY=S%5/valid?[Link].

=se3rge+reIhtX,
f I .find
9 the equation of
YenX regression', where X' = X + 3 and Y = Y + 6.
Rp. r - 9 = 2(X' - 3)

[Link]
110 Statistics

10. When studying the linear regression between average incomes (Yen $) and the number of
children by family (X), the following information was obtained:
x= 3, y - 700,s x =0.5-JsXY,
estimate the income of families with 4 children, how many children per family
Would an estimated income of $712 be appropriate?
RpY= 688 +4X.>' = $704.X=b

11. When estimating sales (10 of an item based on prices (X), a ...
least squares line based on a sample of 4 data points. If sales
observedwere10, 8 , 6 14 and if the respective estimated sales are 10.8.

8.2,5.6,13.4,whatpercentageofthevarianceofsalesisexplainedby
regression line? Rp. 96.57%.

12. In a study of the relationship between monthly income and education expenses
the families, a sample provides a coefficient of determination of
90.25%,respectiveaveragesof$420and$120,andrespectivestandarddeviations
from$10and$[Link]
a) How much are the estimated education expenses of a family whose income
Isthemonthlyamount$300?
b) If a family estimates their spending on education to be $370, how much should it be?
What is your monthly income?

c) If a family has an increase of $50, how much would it increase the


estimation of your expenses in education?
Rp.Y=-159.3+0.6665^. a) $ 40.2, b) $ 795.94, 0.665x50=33.25

13. When studying the relationship between age (X) and blood pressure (10 from)
Asample of women provided the following information:
s x =7.5,s Y = 10,x=50,y= 120, r = 0.90
a) Find the linear relationship of pressure with respect to age and predict it.
blood pressure for a 45-year-old woman.
b) What percentage of the variance of the pressures is explained by the regression of
blood pressure in relation to age?
Rp.a) y=60+l ,2X, si jc=45, y =114. b)81%.

14. When studying the relationship between costs (X) and profits (10 in dollars of
certain products based on a sample obtained the following information:
s x = 5 , s Y=4, x= 100, y= 50, K=-26+0.76X.
a) What percentage of the variance in profits is explained by the regression
on the cost benefits?

[Link]
Simple linear regression 111

b) If each cost value is increased by $3 and the corresponding value to the


utility increases by 6 $, how much is the profit estimated for a cost of

$120?
a) r=0.95. ^=0.9025. el9025%, b) K-56=0.76(A"-1()3), 68.92.

15. Using the n data pairs an equation


. is obtained from
linear regression with a slope of -1. Determine the correlation coefficient
between the values x and y, if additionally, s x/sr = 0.9

Rp.r= -0.9.

16. Sinpares (x, (x„, yn) have correlation index r, verify that
the regression line for the points (xj*,>■*),...,(x*,>'*) where
* x.: - x y : - y..
x, = --------, y¡ = -------- parallel = l, 2,... , / i , esy = rx.
Sx Sy
Rp.x* = 0. v*=0,r*=r,theny'=rx'

17. Pairs (x1, y1), ...(xn, yn) are such that they fulfill the relationship: Y = bX
deduce using the method of least squares. Rp.b='LxyfZx

18. A food company operates a chain of retail stores. For


to measure the efficiency of the stores, the relationship of the number of employees was studied
(X) and the average monthly sales volume (J7) expressed in hundreds of
dollars for all the stores during the last year. The chart of the data
suggest a linear relationship between the variables. The following information is available:

= 100, I X =600, I K =1600, IX K = 13600. I X 2 =5200, I K 2 =37700


a) Find the least squares line to estimate sales based on the number
of employees. What are the estimated sales for a store? 8

Employees?
b) What percentage of the variance in sales is explained by variability?
of the number of employees?
c) How many employees does the store have if the average sale is estimated at $1,100?
Y = l + 2.5X, 21 or $2100.

A market study seeks to find out if the advertising is effective.


televised of a product that went on sale in relation to time of
advertising (in hours/week). Data will be collected from the second
The week of the initiated advertising resulting in the following table. It could not be
collect data from the fourth week.

[Link]
112 Statistics

Week 2 3 4 5 6 7
Propaganda time 2 0 25 22 28 36 40
Sale of Product($ ) 300 310 - 320 350 420

a) Is the product's advertising effective?


b)Howmuchwouldyouestimatethesalesforweek4?
Y=176.82 +5.467X, r=0.92.

20. An editor took a sample of 7 books noting the price and the number of
respective pages, obtaining the following data:
No. of pages 630 550 400 250 370 320 610
Price ($) 10 8 7 4 6 6 9
a) Determine a linear function between the price and the number of pages with it.
end of predicting prices. What percentage of the total variance of prices
Explain this function?
b) Estimate the price of a 300-page book. If this book is increased
2 0 pages in a second edition, how much would the price increase?

c)Howmanypagesshouldabookestimatedtocost$12.27have?
Y = 1.22 + 0.013X; 94.5%

21. A sample of 5 adult males whose heights were observed (X in


feet, inches) and weights (and in pounds) has given the following results:
X 5'2"
5 feet 1 inch 5'3" 5'4" 5'5"
Y 125 130 140 145 160

a) Perform a linear regression and use the data to verify that the variance
Total variance is equal to the residual variance plus the variance explained by the model.
regression line.
b) Using the decomposition of variance, calculate r2 and interpreted it
result.
a) Y = 114.5 + 8.5X', X1 = 1 r.3,4,5. 750 = 27.5 + 722.5. b) r = 0.96.

22. I want to study the relationship between ages in years (X) of a type of
machines used in the manufacturing of a certain item and the number of
articles(y) that produce. Based on the sample from the following table:
a) Determine the least squares regression line to predict the
production. Estimate the production for 4, 7 and 8 years.

b) Calculate the percentage of the variance explained by the regression of the


production.
c) If each machine in the sample really produces 10 fewer items
Determine the regression line. What is the percentage of the variance?
explained by the regression of production?

[Link]
Simple linear regression 113

X Y
2 95
3 70,80
4 -

5 75
6 60
7 -

8 _

9 45, 50
10 25
Y=101.5 - 6.64X 91.5 - 6.64X and R will change.

23. Be the consumer price index, taking 1980 as the base year.
say 1980 = 100). For the following data:
Year 1981 1982 1983 1984 1985 1986 1987
Y 106.0 111.1 117.2 121.3 125.2 128 0 132.6

a) Find the least squares line that fits the data.


b) Predict the price index for the year 1988 and compare it with the value
true (144.4). In what year can we expect the price index
sea 150.57, assuming that the current trends continue?
a) Y=102.83 +4.34X, r= 0.994 b) year 1991.

24. The percentages of advertising expenses and the percentages of net profits
of sales in a sample of 9 businesses as follows:
Expenses 2.3 1.9 3.5 1.0 1.5 4.0 2.6 3.0 2.4
Benefits 4.0 3.8 6.2 2.9 3.4 6.8 4.5 5.0 4.2

a) Find the least squares regression line to predict profits


grandchildren.
b) Determine the profit if the expense is 5%. What is the percentage of the
explained variance of benefits concerning expenses?
Rp.a) K=1.274 + 132X 7.88 and 95.99%.

25. A factory of a certain soft drink brand has randomly taken 9 weeks of the year,
observingtheaveragetemperatureindegreesCelsius(X)and
the quantity of soft drinks in thousands (y) orders during each of those
periods. The data is summarized in the following table:
X 28 14 12 31 30 19 24 15 16
Y 60 19 12 75 70 40 55 25 25

[Link]
114 Statistics

Find the least squares regression line to predict the amount of


Orders. Can production be planned based on temperature?
Y = -23.9 + 3.154*

26. The following data is the selling prices in dollars. And a brand of
used cars years:
X 1 2 3 4 5 6

Y 6350 5695 5790 - 4985 4890

a) Fit a least squares curve of the formY=AB x.


b) Estimate the selling price of a car that has been used for 4 years.
Rp.a) Y = 6559.78(0.95)* ■ r=-0.9658. 3"b)=5
98342.,

27. The following data is the net investment (X) and the interest rate (10
X 12 8 10 7 6 5 5
Y 4 5 6 7 8 9 10

Two models for relating are proposed. YconX

Y=AXb e Y = a + bX

Which model fits the data better? Why?


Non-linear model. l'=40.29(X)_ow)86, r=0.93, ^=0.87, linear model. K=12.63 0.74X í-0.9 1,
r=0.82. Better fits the nonlinear model.

28. Fit a curve of the form by the least squares method:

Y =----- í-----
A+BX
the following pairs of data:
X 4 8 12 16 20 24 28 32
Y 24 21 20 15 14 10 7 5
Rp.r= 4Y. Y'=0.003146+0.005176X, <=0.9127

29. Adjust by the least squares method a curve of the form:

Y = 5 +----- ?-----
A+BX
to the problem data 19. ( do first Y'=Y-5.

30. The pressure P (kg./cm2.) of a gas corresponding to different volumes V


(cm3.) was recorded in the following table:

[Link]

You might also like