<< Chapter < Page Chapter >> Page >

The method of least squares

We now come to a more accurate method of finding the line of best-fit. The method is very simple. Suppose we guess a line of best-fit. Then at at every data point, we find the distance between the data point and the line. If the line fitted the data perfectly, this distance should be zero for all the data points. The worse the fit, the larger the differences. We then square each of these distances, and add them all together.

The best-fit line is then the line that minimises the sum of the squared distances.

Suppose we have a data set of n points { ( x 1 ; y 1 ) , ( x 2 ; y 2 ) , ... , ( x n , y n ) } . We also have a line f ( x ) = m x + c that we are trying to fit to the data. The distance between the first data point and the line, for example, is

distance = y 1 - f ( x ) = y 1 - ( m x + c )

We now square each of these distances and add them together. Lets call this sum S ( m , c ) . Then we have that

S ( m , c ) = ( y 1 - f ( x 1 ) ) 2 + ( y 2 - f ( x 2 ) ) 2 + ... + ( y n - f ( x n ) ) 2 = i = 1 n ( y i - f ( x i ) ) 2

Thus our problem is to find the value of m and c such that S ( m , c ) is minimised. Let us call these minimising values m 0 and c 0 . Then the line of best-fit is f ( x ) = m 0 x + c 0 . We can find m 0 and c 0 using calculus, but it is tricky, and we will just give you the result, which is that

m 0 = n i = 1 n x i y i - i = 1 n x i i = 1 n y i n i = 1 n ( x i ) 2 - i = 1 n x i 2 c 0 = 1 n i = 1 n y i - m 0 n i = 0 n x i = y ¯ - m 0 x ¯

Khan academy video on regression - 1

In the table below, we have the records of the maintenance costs in Rands, compared with the age of the appliance in months. We have data for 5 appliances.

appliance 1 2 3 4 5
age ( x ) 5 10 15 20 30
cost ( y ) 90 140 250 300 380

Find the line of best fit using the method of least squares.

  1. appliance x y x y x 2
    1 5 90 450 25
    2 10 140 1400 100
    3 15 250 3750 225
    4 20 300 6000 400
    5 30 380 11400 900
    Total 80 1160 23000 1650
  2. m 0 = n x y - x y n x 2 - x 2 = 5 × 23000 - 80 × 1160 5 × 1650 - 80 2 = 12 c 0 = y ¯ - b x ¯ = 1160 5 - 12 × 80 5 = 40 y ^ = 40 + 12 x ̲

Using a calculator

Find a regression equation for the following data:

Days ( x ) 1 2 3 4 5
Growth in m ( y ) 1,00 2,50 2,75 3,00 3,50
  1. Using your calculator, change the mode from normal to “Stat x y ”. This mode enables you to type in bivariate data.

  2. Key in the data as follows:

    1 ( x , y ) 1 DATA n = 1
    2 ( x , y ) 2,5 DATA n = 2
    3 ( x , y ) 2,75 DATA n = 3
    4 ( x , y ) 3,0 DATA n = 4
    5 ( x , y ) 3,5 DATA n = 5
  3. Ask for the values of the regression coefficients a and b .

    RCL a gives a = 0,9
    RCL b gives b = 0,55
    y ^ = 0 , 9 + 0 , 55 x

Using a calculator determine the least squares line of best fit for the following data set of marks.

Learner 1 2 3 4 5
Chemistry (%) 52 55 86 71 45
Accounting (%) 48 64 95 79 50

For a Chemistry mark of 65%, what mark does the least squares line predict for Accounting?

  1. Switch on the calculator. Press [MODE] and then select STAT by pressing [2]. The following screen will appear:

    1 1-VAR 2 A + BX
    3 _ + CX 2 4 ln X
    5 e ^ X 6 A . B ^ X
    7 A . X ^ B 8 1/X

    Now press [2] for linear regression. Your screen should look something like this:

    x y
    1
    2
    3
  2. Press [52] and then [=]to enter the first mark under x . Then enter the other values, in the same way, for the x -variable (the Chemistry marks) in the order in which they are given in the data set. Then move the cursor across and up and enter 48 under y opposite 52 in the x -column. Continue to enter the other y -values (the Accounting marks) in order so that they pair off correctly with the corresponding x -values.

    x y
    1 52
    2 55
    3

    Then press [AC]. The screen clears but the data remains stored.

    1: Type 2: Data
    3: Edit 4: Sum
    5: Var 6: MinMax
    7: Reg

    Now press [SHIFT][1]to get the stats computations screen shown below. Choose Regression by pressing [7].

    1: A 2: B
    3: r 4: x ^
    5: y ^
    1. Press [1] and [=]to get the value of the y -intercept, a = - 5 , 065 . . = - 5 , 07 (to 2 d.p.) Finally, to get the slope, use the following key sequence: [SHIFT][1][7][2][=]. The calculator gives b = 1 , 169 . . = 1 , 17 (to 2 d.p.) The equation of the line of regression is thus: y ^ = - 5 , 07 + 1 , 17 x
    2. Press [AC][65][SHIFT][1][7][5][=] This gives a (predicted) Accounting mark of ^ = 70 , 94 = 71 %

Questions & Answers

why we learn economics ? Explain briefly
ayalew Reply
why we learn economics ?
ayalew
why we learn economics
ayalew
profit maximize for monopolistically?
Usman Reply
what kind of demand curve under monopoly?
Mik Reply
what is the difference between inflation and scarcity ?
Abdu Reply
What stops oligopolists from acting together as a monopolist and earning the highest possible level of profits?
Mik
why economics is difficult for 2nd school students.
Siraj Reply
what does mean opportunity cost?
Aster Reply
what is poetive effect of population growth
Solomon Reply
what is inflation
Nasir Reply
what is demand
Eleni
what is economics
IMLAN Reply
economics theory describes individual behavior as the result of a process of optimization under constraints the objective to be reached being determined by
Kalkidan
Economics is a branch of social science that deal with How to wise use of resource ,s
Kassie
need
WARKISA
Economic Needs: In economics, needs are goods or services that are necessary for maintaining a certain standard of living. This includes things like healthcare, education, and transportation.
Kalkidan
What is demand and supply
EMPEROR Reply
deman means?
Alex
what is supply?
Alex
ex play supply?
Alex
Money market is a branch or segment of financial market where short-term debt instruments are traded upon. The instruments in this market includes Treasury bills, Bonds, Commercial Papers, Call money among other.
murana Reply
good
Kayode
what is money market
umar Reply
Examine the distinction between theory of comparative cost Advantage and theory of factor proportion
Fatima Reply
What is inflation
Bright Reply
a general and ongoing rise in the level of prices in an economy
AI-Robot
What are the factors that affect demand for a commodity
Florence Reply
price
Kenu
Got questions? Join the online conversation and get instant answers!
Jobilize.com Reply

Get Jobilize Job Search Mobile App in your pocket Now!

Get it on Google Play Download on the App Store Now




Source:  OpenStax, Siyavula textbooks: grade 12 maths. OpenStax CNX. Aug 03, 2011 Download for free at http://cnx.org/content/col11242/1.2
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'Siyavula textbooks: grade 12 maths' conversation and receive update notifications?

Ask