T~,
~I()
lo.k
~aQw
ARE 157
Spring
2009
HOMEWORK 2
\~
~
t
~
pOSt<..&
I'2I.SW
~
to
1,
Forecasting
using
data
smoothing.
James and
Kelly
are
avid
basketfall
fans.
They often argue about whether players experience "momentum" (hot streaks or
cold
streaks that carry over
from
one
game to
the
next).
James believes momentum
is
important.
If
a player achieves a
high
score
one
night,
he
is likely to perform well again
in
the next
game.
Kelly thinks scoring fluctuates randomly around
the
player's
mean
score,
so
that
knowing
how
much
the player
scored
in
the last few games
does
not help forecast
the
next game's points
scored.
To settle their
dispute,
they look at
2007
data
from
two
players:
Kobe
Bryant,
a leading player with extensive NBA experience;
and
Kevin
Martin, at that time a secondyear
player.
They use
8
data smoothing models to forecast the next game's points scored,
each
implying different degrees
of
momentum
from
one
game to the
next.
a.
Create
the
following
8
sets
of
forecasts, using actual
data
from
Kobe
Bryanfs first
42
games of
the
season.
Begin your forecasts
on
game
6
(games
1
through
5
are needed
to
create some of the forecasts). For your exponential smoothing models, assume your first forecast (Game
6
forecast)
is
28.4,
Kobe's average points
scored
in
the
'0506
season.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '08
 WHITNEY
 exponential smoothing models, 25.4 27.6 10.8 10.8 10.8 10.8 22.2, 5game 3game wtd

Click to edit the document details