Poisson Distribution Goals in English Premier Football League – 2006/2007 Regular Season.
-
Upload
antony-barker -
Category
Documents
-
view
217 -
download
0
Transcript of Poisson Distribution Goals in English Premier Football League – 2006/2007 Regular Season.
Poisson Distribution
Goals in English Premier Football League – 2006/2007 Regular
Season
Poisson Distribution• Distribution often used to model the number of
incidences of some characteristic in time or space:– Arrivals of customers in a queue– Numbers of flaws in a roll of fabric– Number of typos per page of text.
• Distribution obtained as follows:– Break down the “area” into many small “pieces” (n pieces)– Each “piece” can have only 0 or 1 occurrences (p=P(1))– Let =np ≡ Average number of occurrences over “area”– Y ≡ # occurrences in “area” is sum of 0s & 1s over “pieces”– Y ~ Bin(n,p) with p = /n– Take limit of Binomial Distribution as n with p = /n
Poisson Distribution - Derivation
)1,,POISSON( :)(
)0,,POISSON( :)(
:Functions EXCEL
onDistributiy Probabilit "Legitimate" 1!!
)(
! :function lexponentia ofexpansion Series
,...2,1,0!!
)(lim
1lim :get weCalculus, From
1lim!
)(lim
fixed allfor 11
lim...lim :Note
11
...1
lim!
1)(
)1)...(1(lim
!
1)!(
)!)(1)...(1(lim
!1
)!(!
!lim)(lim
: aslimit Taking
1)!(!
!)1(
)!(!
!)(
000
0
yyF
yyp
eey
ey
eyp
i
xe
yy
ee
yyp
en
a
nyyp
yn
yn
n
n
nn
yn
n
n
n
n
ynn
ynnn
y
n
n
nynn
ynynnn
ynnyny
nyp
n
nnyny
npp
yny
nyp
y
y
y
y
y
x
ix
yy
n
an
n
n
n
y
n
nn
n
n
yn
yn
y
yn
yn
yyny
nn
ynyyny
Poisson Distribution - Expectations
2222
22
22
2
22
220
1
1
110
][)()(
)()1(
)!2(
)!2(!)1(
!)1()1(
)!1()!1(!!)(
,...2,1,0!
)(
YEYEYV
YEYYEYE
eey
e
y
e
y
eyy
y
eyyYYE
eey
ey
e
y
ey
y
eyYE
yy
eyf
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
y
Example – English Premier League
• Total Goals Per Game (Both Teams)– Mean=2.47 Variance=2.49
• Goals by Team by Half– Home Team, 1st Half: Mean=0.68 Variance=0.73– Road Team, 1st Half: Mean=0.44 Variance=0.39– Home Team, 2nd Half: Mean=0.77 Variance=0.75– Road Team, 2nd Half: Mean=0.58 Variance=0.83*
*Does not reject based on Goodness-of-Fit test
Goals by Team by Half
Goals All Home1 Road1 Home2 Road20 828 199 236 175 2181 492 121 122 134 1152 157 46 21 56 343 31 11 0 12 84 9 3 1 3 25+ 0 0 0 0 0
Observed Counts
Goals All Home1 Road1 Home2 Road20 818.97 192.72 244.22 175.30 212.991 506.47 130.84 107.97 135.63 123.312 156.60 44.42 23.87 52.47 35.693 32.28 10.05 3.52 13.53 6.894 4.99 1.71 0.39 2.62 1.005+ 0.69 0.26 0.04 0.46 0.13
Expected Counts Under Poisson Model
Goodness of Fit Tests (Lumping 3 and More Together for Team Halves)
Goals Home1 Road1 Home2 Road20 0.2048 0.2766 0.0005 0.11811 0.7407 1.8229 0.0195 0.55972 0.0563 0.3444 0.2381 0.08043+ 0.3263 2.1967 0.1563 0.4928
Chi-Square 1.3282 4.6407 0.4144 1.2509P-value 0.7225 0.2001 0.9373 0.7408
ondistributi- thefollows
statistic square-chi thefits, modelPoisson that thehypothesis null Under the
expected
expected-observed
:by obtained is statistic Square-Chi theon tocontributi thecell,each For
23
22
X
Correlations Among Goals Scored
Correlations Home1 Road1 Home2 Road2Home1 1.0000 0.0491 0.0262 -0.0587Road1 0.0491 1.0000 -0.0388 -0.0475Home2 0.0262 -0.0388 1.0000 -0.0771Road2 -0.0587 -0.0475 -0.0771 1.0000
t-test (r=0) Home1 Road1 Home2 Road2Home1 #N/A 1.0047 0.5239 -1.0774Road1 1.0047 #N/A -0.7259 -0.8808Home2 0.5239 -0.7259 #N/A -1.3910Road2 -1.0774 -0.8808 -1.3910 #N/A
)1,0( tely)(approxima ddistribute is statistic This,0 that hypothesis Under the
21
:0 are ns"correlatio population" he whether tgfor testin Statistic- tThe
2
Nnr
rtobs
r
Observed and Expected Counts - Total Goals Per Game
0
20
40
60
80
100
120
-1 0 1 2 3 4 5 6 7
Goals
Fre
qu
en
cy
Observed
Expected