You are on page 1of 55


 Forecasting is the art and science of predicting future

 Forecast may involve taking historical data and
projecting them into the future with some sort of
mathematical model.
 It may be subjective or intuitive prediction. Or may be
involve a combination of these.
Forecasting Time Horizons

 Short range forecast: This forecast has a time span of up to 1 year

but is generally less than 3 months. It is used for planning purchasing,
job scheduling, workforce levels, job assignment and production
 Medium range forecast: A medium-range forecast generally spans
from 3 months to 3 years. It is useful in sales planning, production
planning and budgeting, cash budgeting and analysis of various
operating plans.
 Long range forecast: Generally 3 years or more in time span, long
range forecasts are used in planning for new products, capital
expenditures, facility location or expansion and research and
Types of Forecast

 Economic forecasts address the business cycle by

predicting inflation rates, money supplies, housing starts and
other planning indicators.
 Technology forecasts are concerned with rate of
technological progress, which can result in the birth of
exciting new product, requiring new plants and equipment.
 Demand Forecasts are projection of demand for a
company’s products or services. These forecasts, also sales
forecasts, drive a company’s production, capacity and
scheduling systems and serves as inputs to financial,
marketing and personnel planning.
Forecasting Approaches

 Quantitative forecasts: use a variety of mathematical

models that rely on historical data.
 Qualitative forecasts: incorporate such factors as the
decisions makers intuition, emotions, personal
experience, and value system in reaching.
Quantitative Methods

 Time Series Models:

 Naïve approach
 Moving average
 Exponential smoothing
 Trend projection
 Associative Model
 Linear regression
Time Series Forecasting
 A forecasting techniques that uses a series of past data points to
make a forecast.
 A time series has four components:
 Trends: the gradual upward or downward movement of data
over time. Changes in income, population, age distribution, or
cultural views may account for movement in trend.
 Seasonality:data pattern that repeats itself after a period of
days, weeks, months or quarters.
 Cycles: Pattern in data that occurs every several years.
 Random variations: ‘blip’ in the data caused by chance and
unusual situation.
Naive Approach

 A forecasting technique which assumes that demand in

the next period is equal to demand in the most recent
Moving Average

 A forecasting method that uses an average of the n

most recent periods of data to forecast the next period.

σ 𝐷𝑒𝑚𝑎𝑛𝑑 𝑓𝑜𝑟 𝑛 𝑝𝑒𝑟𝑖𝑜𝑑

𝑀𝑜𝑣𝑖𝑛𝑔 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 =
Weighted moving average

𝑊𝑒𝑖𝑔ℎ𝑡𝑒𝑑 𝑚𝑜𝑣𝑖𝑛𝑔 𝑎𝑣𝑒𝑟𝑎𝑔𝑒

σ(𝑊𝑒𝑖𝑔ℎ𝑡 𝑓𝑜𝑟 𝑝𝑒𝑟𝑖𝑜𝑑 𝑛)(𝐷𝑒𝑚𝑎𝑛𝑑 𝑖𝑛 𝑝𝑒𝑟𝑖𝑜𝑑 𝑛)
σ 𝑊𝑒𝑖𝑔ℎ𝑡
Problems faced by Moving Average
 Increase the size of n does smooth out fluctuations
better, but it makes the method less sensitive to real
change in data.
 Moving average cannot pick up trend very well.
Because they are average, they are always stay within
past levels and will not predict changes to either higher
or lower levels.
 Moving average requires extensive use of past data.
Exponential Smoothing

 A weighted moving average forecasting technique in

which data points are weighted by an exponential
 𝑁𝑒𝑤 𝐹𝑜𝑟𝑒𝑐𝑎𝑠𝑡 = 𝐿𝑎𝑠𝑡 𝑝𝑒𝑟𝑖𝑜𝑑′ 𝑠 𝑓𝑜𝑟𝑒𝑐𝑎𝑠𝑡 +
𝛼(𝐿𝑎𝑠𝑡 𝑝𝑒𝑟𝑖𝑜𝑑′ 𝑠 𝑎𝑐𝑡𝑢𝑎𝑙 𝑑𝑒𝑚𝑎𝑛𝑑 − Last period′ s forecast
 𝛼 = 𝑠𝑚𝑜𝑜𝑡ℎ𝑖𝑛𝑔 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
 𝐹𝑡 = 𝐹𝑡−1 + 𝛼 𝐴𝑡−1 − 𝐹𝑡−1 = α 𝐴𝑡−1 + (1 − 𝛼)𝐹𝑡−1
 The old forecast for May was 220, and the actual
demand for May was 190. If alpha (𝛼) is 0.15, calculate
the forecast for June. If June demand turns out to be
218, calculate the forecast for July.

June forecast = 0.15(190) + (1 - 0.15)220 = 215.5

July forecast = 0.15(218) + (0.85)215.5 = 215.9
Forecast Error

 Forecast Error=Actual demand – Forecast Value

 3 popular measures for calculating the forecast error are
 Mean Absolute Deviation (MAD)
 Mean Squared Error (MAE)
 Mean Absolute Percent Error (MAPE)
Mean Absolute Deviation

 A measure of the overall forecast error for a model.

σ 𝐴𝑐𝑡𝑢𝑎𝑙 − 𝐹𝑜𝑟𝑒𝑐𝑎𝑠𝑡
Quarter 1 2 3 4 5 6 7 8 9
Demand 180 168 159 175 190 205 180 182
Mean Squared Error

 MSE is the average of the squared differences between

the forecasted and observed values.
σ 𝐹𝑜𝑟𝑒𝑐𝑎𝑠𝑡 𝐸𝑟𝑟𝑜𝑟 2
 Compute the MSE for 𝛼 = 0.1
Mean Absolute Percentage Error

 The average of the absolute differences between the

forecast and actual values, expressed as a percent of
actual values.

σ𝑛𝑖=1 100 𝐴𝑐𝑡𝑢𝑎𝑙𝑖 − 𝐹𝑜𝑟𝑒𝑐𝑎𝑠𝑡𝑖 ൘𝐴𝑐𝑡𝑢𝑎𝑙

Exponential Smoothing with Trend
 Step 1: Compute 𝐹𝑡 , the exponentially smoothed
forecast for period t, using the following equation:
𝐹𝑡 = 𝛼 𝐴𝑡−1 + (1 − 𝛼)(𝐹𝑡−1 + 𝑇𝑡−1 )

 Step 2: Compute the smooth trend, 𝑇𝑡 using the following

𝑇𝑡 = 𝛽 𝐹𝑡 − 𝐹𝑡−1 + (1 − 𝛽)𝑇𝑡−1
 Step 3: Calculate the forecast including trend, 𝐹𝐼𝑇𝑡 =
𝐹𝑡 + 𝑇𝑡
 The monthly sales for Telco Batteries, Inc., were as follows
Trend Projection

 A time-series forecasting method that tits a trends line to

a series of historic data points and then project the line
into the future for forecasts.
𝑦ො = 𝑎 + 𝑏𝑥

𝑦ො =value of the dependent variable

a= y axis intercept
b= slope of the regression line
x= independent variable
σ 𝑥𝑦 − 𝑛𝑥ҧ 𝑦ത
Slope 𝑏 =
σ 𝑥 2 − 𝑛𝑥 2

𝑎 = 𝑦ത − 𝑏𝑥ҧ
Seasonal Variations in Data
If we expected the 2010 annual demand to be 1200 units.
Associative Forecasting Method:
Regression Analysis
 Linear Regression Analysis: A straight-line mathematical
model to describe the functional relationship between
independent and dependent variables.
𝑦ො = 𝑎 + 𝑏𝑥

𝑦ො =value of the dependent variable

a= y axis intercept
b= slope of the regression line
x= independent variable
Standard Error of the Estimate

 A measure of variability around the regression line- its

standard deviation.
σ(𝑦 − 𝑦𝑐 )2
𝑆𝑦,𝑥 =
σ(𝑦 − 𝑦𝑐 )2 σ 𝑦 2 − 𝑎 σ 𝑦 − 𝑏 σ 𝑥𝑦
𝑆𝑦,𝑥 = =
𝑛−2 𝑛−2
Correlation Coefficient for
Regression Lines
 A measure of the strength of the relationship between
two variables.
𝑛 σ 𝑥𝑦 − σ 𝑥 σ 𝑦
[𝑛 σ 𝑥 2 − σ 𝑥 2 ][𝑛 σ 𝑦 2 − σ 𝑦 2 ]
 Room registration in the Toronto Towers Plaza Hotel have
been recorded for the past 9 years. To project future
occupancy, management would like to determine the
mathematical trend of guest registration. This estimate
will help the hotel determine whether future expansion
will be needed. Given the following time series data,
develop a regression equation relating registrations to
time. Then forecast 2011 registrations. Room registrations
are in thousands:
2001:17 2002:16 2003:16 2004:21 2005:20
2006: 20 2007:23 2008: 25 2009: 24
Qualitative Method

 Jury of executive opinion: A forecasting technique that

uses the opinion of a small group of high-level managers
to form a group estimate of demand.
 Delphi Method: A forecasting techniques using a group
process that allows experts to make forecasts
 Sales force composite: A forecasting technique based
on sales persons’ estimates of expected sales.
 Consumer Market Survey: A forecasting method that
solicits input from customers regarding future purchasing