Scribd
Upload
107 views26 pages

Forecasting

This document discusses forecasting methods. It begins by explaining the importance of forecasting for business planning and decision making. It then outlines a 7 step process for developing a forecasting system that includes determining the use and items to forecast, the time horizon, and selecting and validating forecasting models. The document discusses different time horizons for forecasts and types of qualitative and quantitative forecasting methods. It provides details on various quantitative methods like naive method, moving averages, exponential smoothing, and regression analysis. Finally, it covers measuring forecast accuracy.

Uploaded by

RomuloYumul
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
107 views26 pages

Forecasting

This document discusses forecasting methods. It begins by explaining the importance of forecasting for business planning and decision making. It then outlines a 7 step process for developing a forecasting system that includes determining the use and items to forecast, the time horizon, and selecting and validating forecasting models. The document discusses different time horizons for forecasts and types of qualitative and quantitative forecasting methods. It provides details on various quantitative methods like naive method, moving averages, exponential smoothing, and regression analysis. Finally, it covers measuring forecast accuracy.

Uploaded by

RomuloYumul
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 26

FORECASTING

GROUP2

FORECASTING

THE ART AND SCIENCE OF


PREDICTING FUTURE EVENTS.

IMPORTANCE OF FORECASTING
1.
2.
3.

4.
5.
6.
7.
8.

Promotion of new business


Estimation of financial requirements
Smooth and continuous working of a
concern
Correctness of management decisions
Success in business
Plan Formulation
Co-Operation and co-ordination
Complete Control

7 STEPS IN FORECASTING SYSTEM


STEP 1: DETERMINE TE USE OF THE
FORECASTS
STEP 2: SELECT THE ITEMS TO BE
FORECASTS
STEP 3: DETERMINE THE TIME HORIZON
OF THE FORECASTS
STEP4: SELECT THE FORECASTING
MODEL(S)
STEP 5: GATHER THE DATA NEEDED TO
MAKE THE FORECASTS
STEP 6: MAKE THE FORECASTS
STEP 7: VALIDATE AND IMPLEMENT THE
RESULTS.

THE TIME HORIZON

Short Range-Less that 3 months


Short Range forecasts are typically the most accurate.

Medium Range-3 months to 3 years.


Medium Range forecasts are believed to be fairly accurate.
Values of recent activities are already known and trends can
be calculated based on this. Quantitative methods of
forecasting tend to be used very frequently in this time frame.

Long Range-More than 3 years


These long term forecasts can have low accuracy, therefore
requiring constant revisions and updates. The long range time
frame is excellent for planning R&D projects or giant
construction projects that may adapt to changes over time.

TYPES OF FORECASTING
Qualitative Method
Qualitative forecasting techniques are subjective, based
on the opinion and judgment of consumers, experts;
they are appropriate when past data are not available.

Quantitative Method
A statisticaltechniqueformakingprojectionsabout the
future
whichusesnumericalfactsandpriorexperienceto
predict upcomingevents.

QUALITATIVE METHOD

Jury of Executive Opinion


Pool opinions of high-level executives,
sometimes augment by statistical models
Sales Force Composite
estimates from individual salespersons are
reviewed for reasonableness, then
aggregated
Delphi Method
Panel of experts, queried iteratively
Consumer Market Survey
Ask the customer

QUANTITATIVE METHOD

Naive Method

Moving Average

Exponential Smoothing

Trend Method

Regression and Correlation Analysis

NAVE METHOD
Simplest possible forecast
Tomorrow will be like today
Nave si the basis for comparison of all
methods
Ignores any historical data previous today

Year

Actual
A

10

Nave forcast
F

error
E=A-F

2
3
4
5
6
7
8

Year

Actual
A

10

2
3
4
5
6
7

Nave forecast
F

10

error
E=A-F

Year

Actual
A

10

12

Nave forecast
F

error
E=A-F

10

2
3
4
5
6

Use todays result to forecast tomorrow

7
8
N=8

Year

Actual
A

10

1
2
3
4
5
6
7
8
N=8

12
14
15
16
17
19
21
23
?

Nave forecast

error

E=A-F

10
12
14
15
16
17
19
21
23

2
2
1
1
1
2
2
2
SE= 13

MOVING AVERAGE

A forecasting method that uses an average of


the n most recent periods of data forecast the
next period
Determine the moving average

Donnas Garden supply wants a 3 month moving average forecast


including a forecast for next January for shed sales
APPROACH story shed sales are shows in the middle column of the table
below 3 month moving average appears on the right
MONTH
January
February
March
April
May
June
July
August
September
October
November
December

ACTUAL SHED SALES


10
12
13
16
19
23
26
30
28
18
16
14

3 MONTH MOVING AVERAGE

(10+12+13)/3=11 2/3
(12+13+16)/3= 13 2/3
(13+16+19)3=16
(16+19+23)/3= 19 1/3
(19+23+26)/3=22 2/3
(23+26+30)/3=26 1/3
(26+30+28)/3= 28
(30+28=18)/3=25 1/3
(28+18+16)/3=20 2/3

Moving average = demand in previous n periods/ n

REGRESSION AND CORRELATION


ANALYSIS

EXPONENTIAL SMOOTHING

Quantitative forecasting method

Most widely practiced method of time series forecasting

Weighted average of two variables


Ft+1 =

Dt + (1

)Ft

Where
Ft +1 = forecast for next period
Dt

= actual value for present period

Ft = previously determined forecast for present


period

= weighting factor (between 0 and 1)

ADJUSTED EXPONENTIAL
SMOOTHING FORECASTING
METHOD

A method that uses measurable, historical


data observations, to make forecasts by
calculating the weighted average of the
current periods actual value and forecast,
with a trend adjustment added in.

When to Use the Method


Preferred Scenario:
When a trend is present
Good Scenario:
When theres a cyclical or seasonal
pattern

Adjusted Exponential Smoothing:


AFt+1 = Ft+1 + Tt+1
Where
Tt +1 = (Ft+1 Ft ) + (1 - ) Tt
= trend factor for the next period
Tt = trend factor for the current period
= smoothing constant for the adjustment
factor
(just add a trend adjustment factor)
Points to Consider:
To start, pick an unadjusted forecast
In period 1, trend equals 0

Problem: 2005 U.S. Housing Starts (monthly)


Given the following data for 9 months, compute trend
adjusted smoothing average. Use = 0.3 (weighting factor),
= 0.6 (smoothing constant for the trend adjustment factor)
Period
1

Month
Jan

Actual
Demand

Unadjuste
d forecast

Trend

2188

2100

Adjusted
forecast

Feb

2228

2126

16

2142

Mar

1833

2157

25

2182

Apr

2027

2060

-48

2011

May

2041

2050

-25

2025

Jun

2065

2047

-12

2036

Jul

2062

2053

-1

2051

Aug

2038

2055

2056

Sep

2108

2050

-3

2047

Calculations:
Feb : unadjusted forecast:
Ft+1 = Dt + (1 )Ft
= 0.3*2188 + 0.7*2100
= 2126
Trend factor for the next period:
Tt +1 =
(Ft+1 Ft ) + (1 - )Tt
=

0.6*(2126 2100) 0.4*0

16

Trend Adjusted Exponential Smoothing:


AFt+1 = Ft+1 + Tt+1
= 2126 + 16
= 2142

Housing starts

2200

Actual demand
Unadjusted
forecast
Adjusted forecast

2100
2000
1900
1800
Jan

Feb

Mar

Apr

May

Jun

Months

Jul

Aug

Sep

Problem :2 Intel quarterly sales revenue.


Given the following data for 4 months, compute trend
adjusted smoothing average. Use = 0.3 (weighting factor),
= 0.6 (smoothing constant for the trend adjustment factor)

Quarter

Month
ending

Sales revenue
(actual) in $

Unadjusted
forecast(=o.4)
in $

Trend
(=0.7)

Dec-04

110,448

105,000

Mar-05

105,707

Jun-05

115,552

Sep-05

111,396

Dec-05

Adjusted
forecast (AFt) in
$

SOLUTION
Quarter

Month
ending

Sales revenue
(actual) in $

Unadjusted
forecast(=o.4)
in $

Trend
(=0.7)
in $

Adjusted
forecast (AFt) in
$

Dec-04

110,448

105,000

Mar-05

105,707

107,179

1525

108,705

Jun-05

115,552

106,590

45

106,636

Sep-05

111,396

110,175

2523

112,698

Dec-05

110,663

1099

111,762

MEASURES OF FORECAST ACCURACY


Measuring forecast accuracy
MAD = |Actual1 Forecast1|
n
MSE = (Actual1 Forecast1)2
n1
|Actual1- Forecast1| x 100
MAPE =
Actual1
n
MAD Mean absolute deviation
MSE Mean squared error
MAPE Mean absolute percent error
MAD is the average absolute error. MSE is the average of
squared errors and MAPE is the average absolute percent error.

MEASURES OF FORECAST ACCURACY


Period

Actual

Forecast

1
2
3
4
5
6
7
8

217
213
216
210
213
219
216
212

215
216
215
214
211
214
217
216

(A F)

Error
2
-3
1
-4
2
5
-1
-4
-2

|Error|

Error2

2
3
1
4
2
5
1
4
22

4
9
1
16
4
25
1
16
76

Using the figures shown in the


table,
MAD = |e| = 22 = 2.75
n
8
MSE = e2 = 76 = 10.86
n-1 8-1
MAPE = [ |e| x 100 ]
actual
n
= 10.26% = 1.28%

[|Error|
Actual] x
100
.92%
1.41
.46
1.90
.94
2.28
.46
1.89
10.26%

TRACKING SIGNAL

Actual Forecast
Quarter Demand Demand

Error

Absolute Cumulativ
Cumulati
e
Forecast
ve
Absolute
Error
Error
Forecast

Error

1
2

90
95

100
100

-10
-5

-10
-15

10
5

10
15

3
4
5

115
100
125

100
110
110

0
-10
+15

0
-10
+5

15
10
15

30
40
55

140

110

+30

+35

30

85

Mad

Tracking

Signal
(Cumulative
error/mad)

10.0 -10/10= -1
7.5
-15/7.5=
-2
10.0
0/10= 0
10.0 -10/10= -1
11.0
+5/11=
+0.5
14.2 +35/14.2=

+2.5
MAD = | Forecast errors | = 85 = 14.2
n
6
Tracking signal = Cumulative error = 35 = 2.5 MADs
MAD
14.2

SEASONAL INDEX
seasonal index is used when forecasting products
with seasonal demand patterns.
seasonal demand pattern is when a product
experiences a seasonal demand pattern, demand has
a repeatable shape during that timeframe.

Example: Mikki runs a shop and she wishes to determine quarterly seasonal indices
based on her last years sales, which are shown in the table below.

Solution: Using the above formula to find the seasonal index


seasonal index = value of the quarter
quarter average
Find the quarter average
quarter average = 920 + 1085 + 1241 + 446 = 923
4
Find the seasonal index of each season
seasonal index Summer = 920 = 0.997

You might also like