You are on page 1of 3

Fall 2000. Homework #3: Solution.

IE 230

Textbook: D.C. Montgomery and G.C. Runger, Applied Statistics and Probability for Engineers, John
Wiley & Sons, New York, 1999. Chapter 3, Sections 3.3-3.5.

Throughout, define notation and work step by step. Understand the reason for each step. Anyone
quibbling with your solution should be able to point to the specific step in question. Conversely, you
should be able to defend each step.

1. Consider the random experiment whose procedure is to throw one six-sided die. Let the sample
space be S = {1,..., 6}, where the number indicates the side facing up. Assume that all six sides
are equally likely.
(a) Let E 1 = "the outcome is 1". Find P(E 1 ). What result in the textbook did you use?
----------------------------------------------------------------------------
P(E 1 ) = P({1}) = 1/6
"Whenever a sample space consists of n possible outcomes that are
equally likely, the probability of each outcome is 1/n." (M&R, p. 63)
----------------------------------------------------------------------------
(b) Let Ei = "the outcome is i", for i = 1, 2,..., 6. Find P(E 3 ).
----------------------------------------------------------------------------
P(E 3 ) = P({3}) = 1/6.
----------------------------------------------------------------------------
(c) Let E denote the event that the number is less than four. Find P(E).
----------------------------------------------------------------------------
P(E) = P(∪ 3i =1 Ei ) = Σ 3i =1 P(Ei ) = 3×(1/6) = 1/2.
----------------------------------------------------------------------------
(d) Find P(E 2 ∪ E) using Result 3-1 from the textbook. State in words the event.
----------------------------------------------------------------------------
P(E 2 ∪ E) = P(E 2 ) + P(E) − P(E 2 ∩ E) = P(E 2 ) + P(E) − P(E 2 ) = P(E) = 1/2
In words, E 2 ∪ E = "the outcome is a 2 or is less than four",
which is equivalent to "the outcome is less than four",
which is equivalent to "the outcome is 1, 2, or 3".
----------------------------------------------------------------------------

2. Which is the more-general result, Result 3-1 or Result 3-2 in the textbook? (Conversely, which of
these two results is a special case of the other one?)
----------------------------------------------------------------------------
Result 3-1 is more general, with Result 3-2 arising as
the special case when the events are mutually exclusive.
----------------------------------------------------------------------------

- 1 of 3 - Schmeiser
Fall 2000. Homework #3: Solution. IE 230

3. Democratic vice-presidential candidate Joseph I. Lieberman, in clarifying an earlier statement, was


quoted in the September 3, 2000, New York Times as saying
Religion in my opinion, and in my opinion usually is, a source of good behavior.
But two things: I know religious people who I consider not to be moral,
and I also know people who are not religious who I consider to be extremely moral.
So...I am talking here about probabilities.

Let’s try to be precise about these probabilities. Recall that underlying the concept of probability
is the notion of experiment, outcome, sample space, and event. Suppose that the experiment is to
choose a random person from the world population.
(a) In this context, an outcome is ___ a person ______.
(b) In this context, the sample space is S = __ the world population __.
(c) Let R = "the randomly chosen person is religious". Let M = "Mr. Lieberman considers the
randomly chosen person to be moral". Is the event R well defined? Is M? (Recall that an
event is (1) a subset of the sample space and (2) a statement that is either true or false with
each replication of the experiment. Your answer can be based on either meaning, or both.)
----------------------------------------------------------------------------
The event R is not well defined. To be well defined, some test needs to be available to
determine whether the randomly chosen person is religious. In many cases, even the
person and Mr. Lieberman might disagree.
The event M is better defined. Here Mr. Lieberman’s opinion determines whether the
randomly selected person falls within the event.
----------------------------------------------------------------------------
Assume now that R and M are well defined. (For example, think of each person having a label
saying "religious, moral", "religious, not moral", "not religious, moral", or "not religious, not
moral".)
(d) State in words "P(M | R)".
----------------------------------------------------------------------------
P(M | R) is the fraction of world’s religious people who are moral.
----------------------------------------------------------------------------
(e) "I know religious people who I consider not to be moral" can be written as
P(__M′___ | __R__) > 0.
(f) "I also know people who are not religious who I consider to be extremely moral" can be
written as P(__M__ | _R′__) > 0.
(g) "Religion in my opinion, and in my opinion usually is, a source of good behavior" can be
written as P(__M__ | __R__) > P(__M__ | __R′__).
Comment: For a politician, vagueness is often good; for an engineer, vagueness is usually bad.

- 2 of 3 - Schmeiser
Fall 2000. Homework #3: Solution. IE 230

4. (Problem 3-50.) A lot of 100 semiconductor chips contains 20 that are defective. Two chips are
randomly selected, without replacement, from the lot. Let Di = "the ith chip selected is
defective".
(a) What is the probability that the first chip selected is defective?
----------------------------------------------------------------------------
P(D 1 ) = 20 / 100 = .20
----------------------------------------------------------------------------
(b) What is the probability that the second chip selected is defective given that the first one is not
defective?
----------------------------------------------------------------------------
hh
P(D 2 | D 1 ′) = 20 / 99 = 0.202
----------------------------------------------------------------------------
(c) What is the probability that both are defective?
---------------------------------------------------------------------------- hh
P(D 1 ∩ D 2) = P(D 1 ) P(D 2 | D 1 ) = (20 / 100) × (19 / 99) = 0.038
----------------------------------------------------------------------------
(d) Rework (a-c) assuming that sampling is with replacement?
----------------------------------------------------------------------------
Now the second draw behaves just like the first draw.
P(D 1 ) = 20 / 100 = .20
P(D 2 | D 1 ) = 20 / 100 = 0.20
P(D 1 ∩ D 2) = P(D 1 ) P(D 2 | D 1 ) = (20 / 100) × (20 / 100) = 0.04
----------------------------------------------------------------------------

5. (From Problem 3-62.) Laboratory glass is shipped either in small, light packaging or in large,
heavy packaging. Suppose that 2% of the light packages result in broken glass and that 1% is
the analogous number for heavy packages. Suppose that 60% of packages are heavy
(a) Define event notation for broken/unbroken glass and for light/heavy packages.
----------------------------------------------------------------------------
Let B = "a randomly selected package results in broken glass".
Let H = "a randomly selected package is large and heavy".
The complementary events B′ and H′ are then also well defined:
B′ = "a randomly selected package results in unbroken glass".
and H′ = "a randomly selected package is small and light".
----------------------------------------------------------------------------
(b) Write the given probability information in terms of your notation from Part (a).
----------------------------------------------------------------------------
P(H) = 0.6 (and therefore P(H′) = 0.4).
P(B | H) = 0.01 (and therefore P(B′ | H) = 0.99).
P(B | H′) = 0.02 (and therefore P(B′ | H′) = 0.98).
----------------------------------------------------------------------------
(c) What fraction of packages have broken glass?
----------------------------------------------------------------------------
P(B) = P(B | H) P(H) + P(B | H′) P(H′) = (0.01)(0.6) + (0.02)(0.4) = 0.014
----------------------------------------------------------------------------

- 3 of 3 - Schmeiser

You might also like