Professional Documents
Culture Documents
in
Indian scenario
Group 1
21/12/09
Agenda
What is forecasting, Why we do it?
Time horizons
Design and steps
Types
Accuracy and issues
Forecasting in Retail Industry
Forecasting in Petroleum Industry
End note
21/12/09
What is Forecasting ?
Forecasting is the estimation of future
trends by examining and analyzing available
information.
To take effective
Click to edit Master subtitlebusiness
style decisions
• Determine cost to produce a product/service.
• How much money should the company
borrow ?
• When and how borrowed funds should be
repaid ?
• Developing new product/service.
21/12/09
Time horizon in forecasting
Short range (<3 months)
21/12/09
Less 3 months to After 3 years
than 3 3 years
months
21/12/09
Design of forecasting systems
No
Timely
Accurate Reliable
21/12/09
Types of forecasting
Judgmental
Uses subjective inputs
Time series
Uses historical data assuming the future will be
like the past
Associative models
21/12/09
Judgmental
Executive opinions
Sales force composite
Consumer surveys
Outside opinion
Opinions of managers and staff
(Delphi method)
21/12/09
Time series
Trend
Long-term movement in data
Cyclical
Short term variation in data
Seasonality
Caused by unusual circumstances
21/12/09
Components of time series
Irregular
variation
Trend
CyclesCycle
Business
90
89
88
Seasonal variations
21/12/09
Techniques of averaging
Moving average
Exponential
smoothening
21/12/09
Associative models
Regression
Technique for fitting a line to a set of points
21/12/09
Basic concept
Forecast Error :
FE Ai Fi
Where :
Ai = The actual value in time period i
Fi = The forecast value in time period i
21/12/09
Basic concept
Mean Error
1 n
FE [ ( Ai Fi )] Where,
n i 1
FE = Forecast Error
Ai = The actual value in time period i
Fi = The forecast value in time period i
1 n
MSE [ ( Ai Fi )]2
n i 1
Mean Square Error
Where,
21/12/09
MAD = Mean Absolute Deviation
Ai = The actual value in time period i
Fi = The forecast value in time period i
Basic
Mean Absolute Percentage Error
concept
Where,
MAPE = Mean Square Error
Ai = The actual value in time period i
Mean Absolute Deviation Fi = The forecast value in time period i
1 n
MAD Ai Fi Where,
n i 1
MAD = Mean Absolute Deviation
1 n Ai Fi
MAPE 100
n i 1 Ai
Mean Absolute Percentage Error
Where,
21/12/09
MAPE = Mean Square Error
Where,
TS = Tracking signal
Basic concept
SFE = Sum of forecast error
MAD = Mean Absolute Deviation
Tracking Signal
SFE
TS Where,
MAD
TS = Tracking signal
SFE = Sum of forecast error
MAD = Mean Absolute Deviation
21/12/09
Issues while using forecasting
system
21/12/09
Issues while using forecasting
system Click to edit the
Cost outline
Data text format
Availability
Second Outline
Level
How to get Started?
Choice Of Model
Third Outline
Level
Estimation Of Parameters
Fourth
Outline
Level
Time frame Key Inferences from
Fifth
research/practice
Outline
Agency New
reports Click toCompetitor
edit the
outline text format
Second Outline
Issues in using the system:-
Level
How to incorporate external
information Third Outline
Click to edit
Stability
theVs Responsiveness
outline Level
text format
Second Outline Level Fourth
Outline
Third Outline Level
Sales Level
Fourth Outline Level
Promotion Fifth
Fifth Outline Level Outline
When to change the system?
Parameter Re-estimation Vs
Model Change
Forecast
Reliability
21/12/09
Forecasting in Retail sector
21/12/09
Insights to the retail sector
After leading the IT bandwagon, India is poised to grow as a Retail hub
21/12/09
Values in Cr. Rs.
Future forecast of different categories in retail sector
Source: TSMG Analysis
21/12/09
Forecasting - Retail sector
Primarily past two-three years of data and the
targeted growth in the particular good is used to
predict the volume of sales.
21/12/09
Forecasting - Retail sector
High brand retailers Small scale retailers
employ
Click toqualitative
edit the outline
use the forecasting text formaton the
techniques
techniques of the kind historical data and
Second Outline
studied in theory & considering
Level the behavior
forecasting is done for a trends in the market.
longer time duration Third Outline
and involves substantial Level
amounts of asset cost. Fourth
Outline Level
Ex: Daily bazaar
Fifth
Ex: Pantaloons, Big bazaar Outline
21/12/09 Level
Time Period for which forecasting is done?
Example :
For the Diwali season, the forecasting study is
started sometime in March.
21/12/09
Forecasting in petrochemical
industry
21/12/09
Steps - Petrochemical Industry
Detailed production planning for the next year by disintegrating available data
into specific products and scheduling plans.
21/12/09
Previous Scenario
Closure to foreign participation
Sellers’ market
Underdeveloped technology
Result
Limited use of forecasting methods. Industries
believed in ‘Production concept’
21/12/09
Present Scenario
Globalization
Buyers Market
21/12/09
THANK YOU
21/12/09