Scribd
Upload
80 views3 pages

EXE - Forecasting

This document contains two examples analyzing sales data using different forecasting techniques. The first example compares using a 3-month moving average vs a 3-month weighted average to forecast sales. The second example performs a linear regression on sales and advertisements to forecast sales based on a given number of advertisements.

Uploaded by

đà nguyễn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
80 views3 pages

EXE - Forecasting

This document contains two examples analyzing sales data using different forecasting techniques. The first example compares using a 3-month moving average vs a 3-month weighted average to forecast sales. The second example performs a linear regression on sales and advertisements to forecast sales based on a given number of advertisements.

Uploaded by

đà nguyễn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Nguyễn Trung Đà - 11190955

EXE1:

The monthly sales for NW company were as follows:

Month 1 2 3 4 5 6 7 8 9 10

Sales ($m) 22 21 25 27 35 29 33 37 41 37

Forecast November sales using each of the following:

a. Naive method
b. A 3-month moving average
c. A 3 month weighted average using 0.1;0.3;0.6 with the heaviest weights applied to the
most recent months.
d. Which one is better (b or c)?

Forecast November sales using each of the following:

a. Native method: F11(native method) = 37


b. A 3 – month moving average:
10

F11(MA) =
∑ Sales∈month n 37+ 41+ 37
n=8
= =38.3
3 3

c. A 3 month weighted average using 0.1;0.3;0.6 with the heaviest weights applied to the
most recent months.
10

∑ (Sales∈month n) .(weight for month n) 37∗0.1+41∗0.3+37∗0.6


n=8
F11(WMA) = = =38.2
10
0.1+ 0.3+ 0.6
3 ∑ (weight for month n)
n=8

d.

    MA - 3month WMA - 3month


Month Sales Forecating demand /AD/ Forecating demand AD
1 22 - - - -
2 21 - - - -
3 25 - - - -
4 27 F4 = (22+21+25)/3= 22.7 /AD4/=4.3 F4=(22*0.1+21*0.3+ 25*0.6)/1= 23.5 AD4= 3.5
5 35 F5 = (21+25=27)/3= 24.3 /AD5/=10.7 F5= (21*0.1+25*0.3+27*0.6)/1=25.8 AD5= 9.2
6 29 F6 = 29 /AD6/= 0 F6= 31.6 AD6 = -2.6
7 33 F7= 30.3 /AD7/= 2.7 F7=30.6 AD7=2.4
8 37 F8= 32.3 /AD8/ = 4.7 F8=32 AD8=5
9 41 F9= 33 /AD9/ = 8 F9=35 AD9=-2
10 37 F10= 37 /AD10/=0 F10=39 AD10=-2
  SUM: 30.4   17.5

MAD for MA method= 30.4/7 4.3


MAD for WMA medthod = 17.5/7 2.5

Because MAD for MA is smaller than MAD for WMA method => MA is better

EX2: The following data relate the sales and number of advertisements in Maden company:

Month 1 2 3 4 5 6

Number of 3 4 7 6 8 5
avertisements

Sales ($m) 340 610 700 520 1000 767

a. Perform a linear regression that relates sales and number of advertisements


b. If the number of advertisements are 10, what are the sales expected to be?
c. Compute standard error of the estimate and correlation coefficient and interpret the meaning
of those numbers.
a. We have a linear regression that: y = a + bx
With: x is number of avertisements
y is sales

x ( Number of
  avertisements) y (Sales) x^2 x.y
  3 340 9 1020
  4 610 16 2440
  7 700 49 4900
  6 520 36 3120
  8 1000 64 8000
  5 767 25 3835
Sum: 33 3937 199 23315
Average 656.166
: 5.5 7    
b=
∑ xy−n . x . y = 23315−6∗5.5∗656.1667 =95
∑ x 2−n . x 2 199−6∗5.52
a = y−b . x=656.1667−95∗5.5=134
y = 134 + 95x

1200 y (Sales)
1000
800 f(x) = 94.9428571428571 x + 133.980952380952
600
400
200
0
2 3 4 5 6 7 8 9

b. If the number of advertisements are 10, what are the sales expected to be?
y = 134 + 95 * 10 = 1084
c. Standard error of the estimate:

S y , x=

n.
√∑
∑ y 2−a . ∑ y−b . ∑ xy =
n−2
xy −∑ x . ∑ y
√ 2836389−134∗3937−95∗23315 =154
6−2

 r = = 0.6
√¿ ¿ ¿
2
r =0.5

You might also like