Please wait while the transcript is being prepared...
0:04
Examine the situation facing
a manager who must
periodically forecast
the inventories for
hundreds of products.
Each day, week or month,
updated forecasts
for the many products are
required within a short time.
While it might be
possible to develop
sophisticated forecasting models
for each of the items, in
many cases, some very simple
short-term forecasting tools
are adequate for the job.
A manager facing such
a task is likely to
use some form of
time-series smoothing.
All the time-series
smoothing methods use a form
of weighted average of
past observations to
smooth up-and-down
movements, that is,
some statistical method of
suppressing short-term
fluctuations.
The assumption that
the forecaster makes
which underlies these
methods is that
the fluctuations in
past values represent
random departures from some
smooth curve that, once
identified, can plausibly
be extrapolated into
the future to produce a forecast
or series of forecasts.
Each of the time-series
methods examines
different combinations
of the types of
departures from the smooth
curve that commonly occur.
This particular example is
the forecast of a
standard main body kit.
The data is real from a company
that has over 160 plants
located throughout
the United States
supplying the recreational
vehicle industry.
The blue line represents
the sales of these
kits over time.
