0% found this document useful (0 votes)
24 views

Sums and Means - Practice Problems: Pierre Brochu Department of Economics University of Ottawa

This document contains definitions and practice problems related to sums and means. It defines key terms like the mean of variables y and x2. It then lists 3 properties of sums and provides 7 practice problems to show relationships between sums and means. The final section shows the answers to the practice problems by applying the properties of sums defined earlier.

Uploaded by

42度鱼
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
24 views

Sums and Means - Practice Problems: Pierre Brochu Department of Economics University of Ottawa

This document contains definitions and practice problems related to sums and means. It defines key terms like the mean of variables y and x2. It then lists 3 properties of sums and provides 7 practice problems to show relationships between sums and means. The final section shows the answers to the practice problems by applying the properties of sums defined earlier.

Uploaded by

42度鱼
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Sums and Means - Practice problems

Pierre Brochu
Department of Economics
University of Ottawa

Last updated: August 18, 2023

1 Definitions
Assume n draws of the x2 and y variables, i.e. (x12 , y1 ), (x22 , y2 ), . . . , (xn2 , yn ) .

Interpretation:

• y1 is the first draw of the y variable, y2 is the second draw of the y variable, and so on.

• x12 is the first draw of the x2 variable, x22 is the second draw of the x2 variable x2 , and
so on. This means that the first subscript refers to the draw and the second subscript
to the variable name.To hammer the point home, xi2 is the i-th draw of the x2 variable.

Comment: Why call a variable x2 ? Why not simply call it x? This is done as to match the
notation in Greene. In the multiple linear regression model we will have K − 1 explanatory
variables, denoted by x2 , x3 , . . . , xK .1 That means xi2 , xi3 , . . . , xiK are the i-th draws of
x2 , x3 , . . . , xK .

Why not start at x1 ? There is an x1 variable in the multiple (simple) and linear regression
model, but it equals one for all draws, and as such, is not an explanatory variable.

Definitions:

ȳ is the mean of the y variable, i.e.


Pn
i=1 yi
ȳ =
n
1
For the simple linear regression model K = 2 means that we have one explanatory variable (i.e. x2 ).

1
Similarly, x̄2 is the mean of the x2 variable, i.e.
Pn
i=1 xi2
x̄2 =
n

2 Properties of sums
1.
n
X n
X n
X
(xi2 + yi ) = xi2 + yi (1)
i=1 i=1 i=1

2.
n
X n
X
ayi = a yi (2)
i=1 i=1

3.
n
X
a=n·a (3)
i=1

3 Practice Problems
Pn
1. Show that i=1 (yi − ȳ) = 0.
Pn
2. Show that i=1 (xi2 − x̄2 ) = 0.
Pn Pn
3. Show that i=1 (xi2 − x̄2 )(yi − ȳ) = i=1 (xi2 − x̄2 )yi .
Pn
− x̄2 )(xi2 − x̄2 ) = ni=1 (xi2 − x̄2 )xi2 .
P
4. Show that i=1 (xi2

5. Show that ni=1 (xi2 − x̄2 )(yi − ȳ) = [ ni=1 xi2 yi ] − nx̄2 ȳ.
P P

6. Show that if ni=1 ei = 0 then ni=1 (xi2 − x̄2 )(ei − ē) = ni=1 xi2 ei
P P P

7. Show that if ni=1 ei = 0 and ni=1 xi2 ei = 0 then ni=1 (xi2 − x̄2 )(ei − ē) = 0
P P P

2
4 Answers
1.
n
X n
X n
X
(yi − ȳ) = yi − ȳ (by prop. 1) (4)
i=1 i=1 i=1
Xn
= yi − n · ȳ (by prop. 3) (5)
i=1
n P
X
i=1 yi
= yi − n · (by definition of a sum) (6)
i=1
n
Xn X
= yi − yi (7)
i=1 i=1

=0 (8)

2.
n
X n
X n
X
(xi2 − x̄2 ) = xi2 − x̄2 (by prop. 1) (9)
i=1 i=1 i=1
Xn
= xi2 − n · x̄2 (by prop. 3) (10)
i=1
n P
X
i=1 xi2
= xi2 − n · (by definition of a sum) (11)
i=1
n
Xn X
= xi2 − xi2 (12)
i=1 i=1

=0 (13)

3
3.
n
X n
X
[ (xi2 − x̄2 )(yi − ȳ) ] = [ (xi2 − x̄2 )yi − (xi2 − x̄2 )ȳ) ] (14)
i=1 i=1
n
X n
X
= (xi2 − x̄2 )yi − (xi2 − x̄2 )ȳ (by prop. 1) (15)
i=1 i=1
Xn Xn
= (xi2 − x̄2 )yi − ȳ (xi2 − x̄2 ) (by prop. 2) (16)
i=1 i=1
Xn
= (xi2 − x̄2 )yi − ȳ · 0 (17)
i=1
n
X
= (xi2 − x̄2 )yi (18)
i=1

4.
n
X n
X
[ (xi2 − x̄2 )(xi2 − x̄2 ) ] = [ (xi2 − x̄2 )xi2 − (xi2 − x̄2 )x̄2 ) ] (19)
i=1 i=1
Xn n
X
= (xi2 − x̄2 )xi2 − (xi2 − x̄2 )x̄2 (by prop. 1) (20)
i=1 i=1
n
X n
X
= (xi2 − x̄2 )xi2 − x̄2 (xi2 − x̄2 ) (by prop. 2) (21)
i=1 i=1
Xn
= (xi2 − x̄2 )xi2 − x̄2 · 0 (22)
i=1
Xn
= (xi2 − x̄2 )xi2 (23)
i=1

4
5.
n
X
(xi2 − x̄2 )(yi − ȳ) (24)
i=1
n
X
= [ xi2 yi − xi2 ȳ − x̄2 yi + x̄2 ȳ ] (25)
i=1
Xn n
X n
X n
X
= xi2 yi − xi2 ȳ − x̄2 yi + x̄2 ȳ (by prop. 1) (26)
i=1 i=1 i=1 i=1
Xn Xn n
X
= xi2 yi − ȳ xi2 − x̄2 yi + n · x̄2 ȳ (by prop. 2 and 3) (27)
i=1 i=1 i=1
n Pn Pn
X
i=1 xi2 i=1 yi
= xi2 yi − n · ȳ − n · x̄2 + nx̄2 ȳ (28)
i=1
n n
n
!
X
= xi2 yi − nȳx̄2 − nx̄2 ȳ + nx̄2 ȳ (29)
i=1
n
!
X
= xi2 yi − nx̄2 ȳ (30)
i=1

6.
n
X
(xi2 − x̄2 )(ei − ē) (31)
i=1
n
X
= xi2 ei − nx̄2 ē (32)
i=1
n P2
i=1 ei
X
= xi2 ei − nx̄2 · (33)
i=1
n
n
X 0
= xi2 ei − n · x̄2 · (34)
i=1
n
Xn
= xi2 ei (35)
i=1

5
7.
n
X
(xi2 − x̄2 )(ei − ē) (36)
i=1
n
X
= xi2 ei − nx̄2 ē (37)
i=1
n P2
i=1 ei
X
= xi2 ei − n · x̄2 · (38)
i=1
n
0
= 0 − n · x̄2 · (39)
n
=0 (40)

You might also like