Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
36 views

Lecture 7

Company A claims its average time to replace four tires is 65 minutes. To verify this, a quality auditor collected service times for 100 customers totaling 6832 minutes. The largest time was 312 minutes. The document discusses using this data to evaluate the company's claim under the assumptions that service times are exponentially distributed and asks the reader to consider: (1) whether a 312 minute service time invalidates the claim; (2) if the total time of 6832 minutes agrees with the theoretical distribution; and (3) what decision the auditor should make about accepting or rejecting the claim. It also discusses insurance for beachfront properties and calculating: (1) the probability of a homeowner experiencing a hurricane during a 30

Uploaded by

Anonymous T02GVGzB
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
36 views

Lecture 7

Company A claims its average time to replace four tires is 65 minutes. To verify this, a quality auditor collected service times for 100 customers totaling 6832 minutes. The largest time was 312 minutes. The document discusses using this data to evaluate the company's claim under the assumptions that service times are exponentially distributed and asks the reader to consider: (1) whether a 312 minute service time invalidates the claim; (2) if the total time of 6832 minutes agrees with the theoretical distribution; and (3) what decision the auditor should make about accepting or rejecting the claim. It also discusses insurance for beachfront properties and calculating: (1) the probability of a homeowner experiencing a hurricane during a 30

Uploaded by

Anonymous T02GVGzB
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Lecture February, 22, 2012

Case studies to illustrate use of some common distributions


1. Company A is in the business of auto repairs, offering express services.
The star service is replacement of four tires, which the company claims
it is done in time averaging to 65 minutes. When company A applies
for a quality assurance certificate, you are selected to perform a quality
audit and, among other things, verify their 65-minute mean service time
claim.
You know from experience (and company A is willing to confirm) that
service time is exponentially distributed. Unable to check all times for
all customers, you collect at random one hundred service times, and find
that the total adds up to 6832 minutes. Further, the largest happens to
be 312 minutes. With this evidence on hand, you must decide whether
the claimed 65-minute mean service time is plausible or not.
(a) If service time is indeed exponentially distributed, exp(), with
m = 65 minutes, what is ?
(b) With said , how likely is a service time as large, or larger, as
312 minutes? Do yo think this large service time invalidates the
companys claim?
(c) Always with as in (a), what would be the probability of no
times of 312 minutes, or above, among the one hundred times
examined? Do yo think now a 312-minute service time invalidates
the companys claim?
(d) Always with as in (a), what is the distribution of the total service
time for 100 customers?
(e) Is the observed total time of 6832 minutes in good agreement with
the theoretical distribution in (1d)?
(f) What would be your decision? Would you accept the company
claim, or reject it as implausible given the available evidence?
2. Coastal Insurance Company underwrites insurance for beachfront properties along the Virginia, Carolina and Georgia coasts. It uses the
estimate that the probability of a Category III hurricane (sustained
winds of more than 110 mph or higher) striking a location of the coast
in any one year is 0.05. We will assume the probability of more than
one is negligible.
1

(a) What is the probability that a homeowner purchasing a property


with a 30-year mortgage will experience at least one hurricane
of Category III during the mortgage period? Use both the binomial and Poisson approximation approach. Would a normal
approximation be adequate here?
(b) If Coastal Insurance Company contracts insurance for 1,000 homeowners, in a fairly small area, what is the expected number of
claims per year caused by hurricanes of Category III?
(c) What is the distribution of the number of claims caused by Category
III hurricanes? (Remember all homeowners are in a fairly small
area.)
(adapted from Lind et al. (2011))

Things to investigate on your own.


1. Case 1 makes some assumptions (and raise some questions) that deserve
further thought.
(a) Service times are assumed to be exponentially distributed. This is
a rather bizarre assumption, in view of the memoryless property
exponential.
(b) Two possible ways of deciding about the companys claim are
sketched: one looks only at the largest service time, other at the
sum of service times. In both cases, we attempt to assess whether
they are plausible given the companys claim.
This raises some questions: which of the two methods is better?
Is there something still better than either? In both cases we are
looking at a summary of the sample: the largest and the sum of
the service times. Would looking at each service time individually
provide some more information?
You will see systematic treatment of these questions in a few weeks,
as we deal with so-called sufficient statistics. For now, squeeze
your intuition, see what it tells you.

References
D. A. Lind, W. G. Marchal, and S. A. Wathen. Statistical Techniques in
Business and Economics. McGraw Hill Higher Education, 2011.
2

You might also like