You are on page 1of 25

Application of Forecasting

Techniques to Forecast
Sales of a Apparel Retailer
Name ID

Asraful Alam 1531210660

Ananya Haq 1530283660

Mayeesha Fairuz 1611529660


Rahman
Ayesha Akhter 1310434060

Md. Rakibul Hussain 1530577660


Shah
Forecasting
 A planning tool that helps management in its attempts to cope
with the uncertainty of the future
 Relies mainly on data from the past and present and analysis of
trends.

 Steps to follow while forecasting:


• Decide on the purpose of the forecast
• Fix a time frame
• Choose a forecasting technique
• Collect and scrutinize related data/information
• Prepare the forecast
• Evaluation of actual performance against that has been forecasted
Company Background

 A Social Business brand of Japan as well as the top most brand of


Asia in the sector of apparels
 Started its operation in Bangladesh on September 2010 and on
2013 its first store had been opened in Dhaka city
 Has 9 stores operating in Bangladesh
 Main category of products involves polo t-shirt, ladies kurti, jeans,
jackets, shirts, kids outfits etc.
 Highly concerned of solving social issues of Bangladesh through
business
Objectives of the Report

 Trend analysis of the base sales data of year 2014 to 2016


 Identification of differences (if any) of the trends between year
2014 to 2016
 Application and analysis of three major averaging techniques of
forecasting
 Application of regression analysis to see how it fits with our data
 Application of different measures of accuracy test
 Identification of the forecasting techniques that give us the best
possible forecast
Forecasting Techniques

Judgment and Opinion Averaging Techniques


 Executive Opinion  Simple Moving
 Sales-Force Average
Opinion  Weighted Moving
 Consumer Surveys average
 Delphi Method  Exponential
Smoothing

Associative Forecasting
Techniques
 Simple & Multiple
Linear Regression
Analysis
Approaches to Evaluate

 Mean Absolute Deviation (MAD)


 Mean Squared Errors (MSE)
 Mean Absolute Percentage Error (MAPE)
Methodology
Topic Selection

Data Accumulation

Date Organization and Time set

Selection of Forecasting Techniques


• Simple Moving Average

• Weighted Moving Average

• Exponential Smoothing

• Regression Analysis (Simple Linear and Multiple Regression)


Methodology

Software – Microsoft Excel

Evaluation Techniques
• Mean Absolute Deviation (MAD)

• Mean Squared Errors (MSE)

• Mean Absolute Percentage Error (MAPE)


Base Data Set
Actual Sales
Month Year
2013 2014 2015 2016
January   1790290 4412739 9202318
February   3481990 2351934 10391970
March   4384980 6063970 11603831
April   4631650 7491518 13931570
May   4883260 8579014 13000314
June   5587310 14961309 43960083
July 3054170 20807900 35515042 24430419
August 2452290 4953944 9215456 14158469
September 833160 7920766 15826092 19774524
October 2497690 8098074 9168621 15968109
November 1236550 5651363 8858105 15014633
December 2117540 8866198 15385103 21033111
Trend Analysis -Monthly
Trend 2014
Monthly Sales of 2014
25000000

20000000

15000000

10000000

5000000

0
Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14 Oct-14 Nov-14 Dec-14
Trend Analysis -Monthly
Trend 2015
Monthly Sales of 2015
40000000

35000000

30000000

25000000

20000000

15000000

10000000

5000000

0
Jan-15 Feb-15 Mar-15 Apr-15 May-15 Jun-15 Jul-15 Aug-15 Sep-15 Oct-15 Nov-15 Dec-15
Trend Analysis -Monthly
Trend 2016
Monthly Sales of 2016
50000000

45000000

40000000

35000000

30000000

25000000

20000000

15000000

10000000

5000000

0
Jan-16 Feb-16 Mar-16 Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16
Trend analysis – 3 year’s
Monthly sales
Monthly Sales (2014-2016)
50000000
45000000
40000000
35000000
30000000
25000000
20000000
15000000
10000000
5000000
0
ry ry ch ril ay ne ly st r e r r r
ua u a ar Ap M Ju Ju g u be ob be be
br m t em m
J an M Au te c v e
Fe p O o ec
Se N D

Sales of 2014 Sales of 2015 Sales of 2016


Eid Sales vs. Other Sales

Percentage of Eid Sales and


Other sales

66.66%

33.34%

Eid Sales Other Sales


Analysis – Simple Moving
Average
Three-Period Moving Average Six-Period Moving Average
50000000 50000000
45000000 45000000
40000000 40000000
35000000 35000000
30000000 30000000
25000000 25000000
20000000 20000000
15000000
15000000
10000000
10000000
5000000
5000000
0
0
Jul-14

Jul-15

Jul-16
Jan-15

Jan-16
Jan-14
Mar-14

Sep-14
Nov-14

Mar-15

Sep-15
Nov-15

Mar-16

Sep-16
May-14

May-15

May-16

Nov-16

4 4 4 4 5 5 5 5 6 6 6 6
n-1 r-1 l-1 t-1 n-1 r-1 l-1 t-1 n-1 r-1 l-1 t-1
Ja Ap Ju Oc Ja Ap Ju Oc Ja Ap Ju Oc

Actual Forecast Actual Forecast


Analysis – Weighted Moving
Average
Weighted Moving Average Weighted Moving Average (with Consideration of Eid Sales Month)
50000000 50000000
45000000 45000000
40000000 40000000
35000000 35000000
30000000 30000000
25000000 25000000
20000000 20000000
15000000 15000000
10000000 10000000
5000000 5000000
0 0
Ap 14

Ju 4
O 14
Ja 14
Ap 15

Ju 5
O 15
Ja 15
Ap 16

Ju 6
O 16
6
r-1
r-1

r-1

-1

6
4

6
1

1
4

6
5
l-

l-
n-

-
n-

l-

-
n-

6
5
1

1
1

1
1
ct

ct

ct

1
l-
l-

l-
Ju

Ju

Ju
n

n
-

-
-

-
-
r

r
-

-
p

p
p
Ja

Ja
Ja

Ja
t

t
c

c
A

A
O

O
Actual Forecast
Actual Forecast
Analysis – Exponential
Smoothing
Exponential Smoothing
50000000
45000000

40000000

35000000

30000000

25000000

20000000

15000000

10000000

5000000

0
Mar-16

Apr-16

May-16

Jul-16

Aug-16

Sep-16

Oct-16

Nov-16

Dec-16
Jan-16

Feb-16

Jun-16

Actual Forecast
Analysis – Simple Linear
Regression
Simple Linear Regression
Actual vs. Forecast
50000000
45000000
40000000
35000000
30000000
25000000
20000000
15000000
10000000
5000000
0
Mar-16

Jun-16

Dec-16
Jan-16

Feb-16

Apr-16

May-16

Jul-16

Aug-16

Sep-16

Oct-16

Nov-16
Actual Forecast
Analysis – Multiple Linear
Regression
Multiple Regression Actual vs. Forecast
50000000

45000000

40000000

35000000

30000000

25000000

20000000

15000000

10000000

5000000

0
Jan-16

Aug-16

Dec-16
Feb-16

Mar-16

Apr-16

May-16

Jun-16

Jul-16

Sep-16

Oct-16

Nov-16
Actual Forecast
Evaluation of Averaging
Techniques
(MAD, MSE, MAPE)
Averaging Techniques
Evaluation of Averaging Techniques
MAD MSE MAPE
Three-Period Moving Average
(From Jan-14 to Dec-16 43.18
Cumulatively) 5148863.815 73297427599013.70 %
Six-Period Moving Average
(From Jan-14 to Dec-16 40.20
Cumulatively) 4793810.972 73121343223831.90 %
Weighted Moving Average
(From Jan-14 to Dec-16 44.48
Cumulatively) 5173609.49 73329223965354.60 %
Weighted Moving Average (with
Consideration of Eid Sales)
(From Jan-14 to Dec-16 43.95
Cumulatively) 4793284.856 51772142124998.50 %
Exponential Smoothing (Only 27.14
from Jan 16 to Dec 16) 5696478.857 109937797469110.00 %
Evaluation of Associative Forecasting
Techniques (MAD, MSE, MAPE)

Evaluation of Associative Forecasting Techniques

Associative Techniques Test Measures 2015


MAD 5769139.619
Simple Linear
Regression MSE 81977530300424.80
MAPE 31.73%
MAD 3331413.875

Multiple Regression
MSE 55819826799313.90
MAPE 12.22%
Findings and Conclusion
 Simple moving average

• Indicated that applying equal weights can lead to somewhat over


and under stocking

 Exponential smoothing

• Was able to somewhat accurately reflect the trends

• Could not accurately predict the timing of the demand with a


detrimental lagging effect on the forecasts
Findings and Conclusion

 Weighted moving average with the consideration of Eid sales


months

• Showed better forecasting to some extent

• Can become a very good forecasting technique to use in future with


company’s maturity in the market

 With the lowest MAD and MAPE, the multiple regression method
has proved to be a better technique

You might also like