C E E 5 9 7 – R i s k A n a l y s i s a n d M a n a g e m e n t
March 5, 2007
Risk Analyses in Transportation Systems (using statistics)
Barnett, Abraham and Schimmel
Airline Safety: Empirical Findings
Barnett and Higgins
Airline Safety: The Last Decade
Flying & Driving
See also:
AirDisaster.Com, AirSafe.Com
Please read for Monday 3/12/07:
National Research Council, "55: A Decade of Experience," Exec. Summary
How should
λ
of Poisson Process be estimated?
How accurate is an estimate?
Method of moments estimator
Over time period T observe K arrivals.
E[K] =
λ
T. Hence if observe that K equals k in this experiment,
set k =
ˆ
!
T, and
obtain
ˆ
= k/T.
Bias of the estimator
ˆ
= K/T
E[
ˆ
] = E[
K/T
] = E[
K
]
/T
=
λ
T /T =
λ
=> Estimator unbiased!!!
Variance of the estimator
ˆ
= K/T
Var[
ˆ
] = Var[
K/T
] = E[
K
]
/T
2
=
λ
T /T
2
=
λ
/T
=> variance of estimator decreases as 1/T with the time we watched
County’s Infant death rate soars ….
reported the headline of the16 Jan. 1990
Ithaca Journal
, that went on to report that
“… health experts who are shocked by the numbers are trying to find an
explanation. Fourteen of every 1,000 babies born in the county during 1987 died
probably before they could walk on their own.”
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentC E E 5 9 7 – R i s k A n a l y s i s a n d M a n a g e m e n t
March 5, 2007
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '07
 Stedinger
 Normal Distribution, Probability theory, Exponential distribution, Poisson process

Click to edit the document details