Faculty of Financial & Administrative Sciences
B y: Dr. Ola E lgeuoshy
S pring 2013
C hapter (3)
“ a Statement about the future value of a variable of
i nterest .”
U ses of Forecasting:
Cash flow and funding
Pricing, promotion, strategy
IT/IS systems, services
Schedules, MRP, workloads
New products and services
F EATURES COMMON TO ALL FORECASTS
Assumes causal system
p ast ==> future
Forecasts rarely perfect because of
Forecasts more accurate for
g roups vs. individuals
Forecast accuracy decreases
a s time horizon increases
I see that you will
get an A this semester.
E LEMENTS OF A GOOD FORECAST
S TEPS IN THE FORECASTING PROCESS
Step 6 Monitor the forecast
Step 5 Prepare the forecast
Step 4 Gather and analyze data
Step 3 Select a forecasting technique
Step 2 Establish a time horizon
Step 1 Determine purpose of forecast
A PPROACHES TO FORECASTING
Q ualitative methods:
c onsist mainly of s ubjective i nputs, which often d efy
p recise numerical description .
I nvolve either the projection of h istorical data o r the d evelopment of a ssociative models t hat attempt to
u tilize c asual (explanatory ) variables to make a
Quantitative techniques p ermit inclusion of soft
i nformation ( e.g.: human factors, personal opinions,
h unches) in the forecasting process.
F ORECASTING TECHNIQUES
F orecasts that use s ubjective inputs s uch as opinions
f rom consumer surveys, sales staff, managers,
e xecutives and experts.
T ime series forecasts:
F orecasts that p roject patterns i dentified in recent
t ime series observations.
F orecasting technique that uses e xplanatory variables
t o predict future demand.
T IME SERIES FORECASTS
Time series: a t ime - o rdered sequence of observations t aken a regular intervals.
The analysis of time series data requires the analyst
t o identify the u nderlying behavior o f the series .
Trend - l ong - term movement in data
Seasonality - s hort - term regular variations in data
Cycle – w avelike variations of more than one year’s d uration
Irregular v ariations - c aused by unusual circumstances Random v ariations - c aused by chance
F ORECAST VARIATIONS
N ATIVE FORECAST
Native forecast i s a forecast for any period that equals the p revious period`s actual value.
D isadvantage i s a ccuracy issue.
Simple to use
Virtually no cost
Quick and easy to prepare
Data analysis is nonexistent
Cannot provide high accuracy
Can be a standard for accuracy
T ECHNIQUES FOR AVERAGING
1 . Moving Average: t echnique that average a number of recent actual v alues, updated as new values become available.
( A) Compute a t hree period m oving average forecast g iven demand for s hopping car ts for t he last five periods .
( B) (B) If t he actual demand in period 6 t urns to be 38, the m oving a verage forecast for period 7 would be
The 3 most recent demands
( A) F 6 = ( 43+40+41)/3= 41 .33
( B) F 7 = (40+41+38)/3= 3 9.67
T ECHNIQUES FOR AVERAGING
2 . Weighted M oving A verage: m ore recent values in a series are g iven more weight in computing a forecast.
G iven the following demand data,
a . Compute a...
Please join StudyMode to read the full document