Professional Documents
Culture Documents
Variance Estimation
Variance Estimation
Variance Estimation
=
2
1
2
1, 1
1
( ) 1 2
(1 ) ( )
1
n
n
X j
j
S X j
E Cov X X
n n n n
o
+
=
( (
=
( (
1
2
1, 1
1
1
( ) 2 (1 ) ( )
n
n X j
j
j
Var X Cov X X
n n
o
+
=
(
= +
(
12
Output Analysis: Variance Estimation IE680
Stochastic Stationary Process (3)
The expected value of the variance estimator is:
If Xi are independent, then is an unbiased estimator of
If the autocorrelation is positive, then is biased low as an estimator of
If the autocorrelation is negative, then is biased high as an estimator of
2
2
1
( )
( )
1
( )
( )
n n
n
n
n
n
S X a
E Var X
n n
S X
E Var X when positively correlated
n
(
=
(
(
<<
(
( )
n
Var X
( )
n
Var X
( )
n
Var X
2
( )
n
S X
n
2
( )
n
S X
n
2
( )
n
S X
n
13
Output Analysis: Variance Estimation IE680
Replication Method
Use to estimate point-estimator variability and to construct a confidence
interval.
Approach: make R replications, initializing and deleting from each one the
same way.
Important to do a thorough job of investigating the initial-condition bias:
Bias is not affected by the number of replications, instead, it is affected only by
deleting more data or extending the length of each run (i.e. increasing T
E
).
Basic raw output data {X
rj
, r = 1, ..., R; j = 1, , n} is derived by:
Individual observation from within replication r.
Batch mean from within replication r of some number of discrete-time observations.
Batch mean of a continuous-time process over time interval j.
14
Output Analysis: Variance Estimation IE680
Replication Method
Length of each replication (n) beyond deletion point (d):
(n - d) > 10d
Number of replications (R) should be as many as time
permits, up to about 25 replications.
For a fixed total sample size (n),
as fewer data are deleted (d decreases)
C.I. shifts: greater bias.
Decrease Variance
Trade off between bias and variance.
15
Output Analysis: Variance Estimation IE680
Batch means method
When using a single, long replication:
Problem :
Data are dependent so the usual estimator is biased
Replication method (more than 25 replications) is expensive
One solution: Batch means method
16
Output Analysis: Variance Estimation IE680
Non-overlapping batch mean (NBM)
m
observations
with batch
mean Y
1,m
Batch 1
m
observations
with batch
mean Y
k,m
Batch k
Batch means method (Nonoverlapping batch mean)
1,
2, , ,
1 1 2 , ( 1) 1 , ( 1) 1
,..., , ,..., ... ,..., ... ,...,
m
m i m k m
n k m
m m m i m im k m km
Y
Y Y Y
X
X X X X X X X X
=
+ + +
17
Output Analysis: Variance Estimation IE680
Nonoverlapping batch mean (NBM)
Suppose the batch means become uncorrelated
as m
NBM estimator for
2
Confidence Interval
,
,
( )
( ) ( )
i m
n i m
nVar Y
nVar X mVar Y
k
~ =
2
,
1
( , ) ( )
1
k
B i m n
i
m
V k m Y X
k
=
=
1,1 / 2
( , )
B
n k
V k m
X t
n
o
18
Output Analysis: Variance Estimation IE680
Overlapping Batch Mean (OBM)
OBM estimator for
2
1 2 3 , 1 2 3
, , , ... , , , , ...
m m m m
X X X X X X X
+ + +
Y
1,m
Y
2,m
1
2
,
1
( ) ( )
( 1)( )
n m
O i m n
i
nm
V m Y X
n m n m
+
=
=
+
19
Output Analysis: Variance Estimation IE680
NBM vs. OBM
Under mild conditions
Thus, both have similar bias
Variance of the estimators
Thus, the OBM method gives better MSE
2
( , ) ( 1) (1 )
B
E V k m k n o n o
(
~ + + +
2
( ) (1 )
O
E V m m o m o
(
~ + +
( , )
3
, ,
2
( )
B
O
Var V k m
as k m
Var V m
(
(
20
Output Analysis: Variance Estimation IE680
Standardized Time Series
Define the centered partial sums of X
i
as
Central Limit Theorem
Define the continuous time process
Question: How does Z
n
(t) behave as n increases?
21
Output Analysis: Variance Estimation IE680
n=100
22
Output Analysis: Variance Estimation IE680
n=1000
23
Output Analysis: Variance Estimation IE680
n=10000
24
Output Analysis: Variance Estimation IE680
n=1,000,000
25
Output Analysis: Variance Estimation IE680
Functional Central Limit Theorem (FCLT)
where B(t) is the standard Brownian motion
(drift coefficient =0, diffusion coefficient =1)
Brownian motion with drift coefficient o and diffusion coefficient
o
2
is a real valued stochastic process with stationary and
independent increments having continuous sample paths where
Thus,
1
( ) ( ) (0 1)
nt
i
i
n
X nt
Z t B t as n t
n
o
(
=
(
s s
2
2 1 2 1 2 1
( ) ( ) ~ ( ( ), ( ))
( ) ( ) (0) ~ (0, )
n n n
B t B t Normal t t t t
Z t Z t Z Normal t
o o
=
26
Output Analysis: Variance Estimation IE680
Functional Central Limit Theorem (FCLT)
While CLT says that for any t,
FCLT also shows that Z
n
(t) converges to an
asymptotically continuous stochastic process with
independent increments.
27
Output Analysis: Variance Estimation IE680
The Weighted Area Estimator
If we can define the function f such that
Continuous [0,1]
Normalized so that
Then,
Define
( )
1
0
( ) ( ) 1 Var f t B t dt =
}
1
0
( ) ( ) ~ (0,1) f t B t dt N
}
( )
2
1
2 2 2
1
0
( ) ( ) ( ) ~ ( ) A f f t B t dt E A f o o _ o
(
=
(
}
28
Output Analysis: Variance Estimation IE680
The Weighted Area Estimator
Since
and
is a variance estimator
1
( )
nt
i
i
n
X nt
Z t
n
o
(
=
(
2
1
0
2
1
( ; ) ( ) ( ) ( ) ( ) ( )
1
n
i i
A f n f Z A f f t B t dt as n
n
n n n
i
o o
(
(
~
(
(
(
=
}
( ) ( )
2 2 2 2
1
( ; ) ( ) ~ ( ) ( )
d
A f n A f E A f E A f o _ o o = =
( ; ) A f n
29
Output Analysis: Variance Estimation IE680
Additional Methods
Regenerative Method (Fishman 1973; Shedler 1993)
Spectral Theory (Heidelberger and Welch 1981, 1983;
Damerdji 1991)
Quantile estimation (Heidelberger and Lewis 1984;
Seila 1982)
Estimation of functions of means (Munoz and Glynn
1997)
Estimation of multivariate means (Anderson 1984;
Charnes 1989, 1995; Seila 1984)
30
Output Analysis: Variance Estimation IE680
Reference
Alexopoulos, C., D. Goldsman, and R. F. Serfozo. 2005. Stationary processes: statistical
estimation. In The Encyclopedia of Statistical Sciences, ed. N. L. Johnson, and C. B. Read,
2nd edition. New York: John Wiley & Sons
Kim, S.-H., C. Alexopoulos, K.-L. Tsui, and J. R. Wilson. 2006. A distribution-free tabular
CUSUM chart for autocorrelated data. IIE Transactions, forthcoming.
Alexopoulos, C., N. T. Argon, D. Goldsman, and G. Tokol. 2004. Overlapping variance
estimators for simulations. Technical Report, School of Industrial and Systems Engineering,
Georgia Institute of Technology, Atlanta, Georgia.
Law, A. M., and W. D. Kelton. 2000 Simulation Modeling and Analysis, 3rd ed. New York:
McGraw-Hill.
31