Professor:
John Millson
Office phone
301-405-5156
Class time:
Tu-Th
11:00-12:15
Class location:
Armory 0126
Office hours:
after class on Thursday
e-mail address:
jjmillson@gmail.com
Web page:
www.math.umd.edu/~jjm
Course text:
Probability and Statistics for Engineering and the
Sciences, University of Maryland, Ninth Edition by Jay Devore
TEACHING ASSISTANTS
Ying Li
Office: CSS Atlantic Building 4326
Sections
0211,0221,0231 (end in 1)
Office hours Monday 10:00 - 11:00
email:yli42@math.umd.edu
Haeyun Seo
Office: MTH 3301
Sections
0212,0222,0232 (end in 2)
Office hours Wednesday 1:00 - 2:00
email:hys2500@math.umd.edu
Manyuan Tao
Office MTH 0203
Sections
0213,0223,0233 (end in 3)
Office hours Monday 10:00 to 11:00
email:mtao1@math.umd.edu
HW12
SOME MORE EXTRA PROBLEMS TO BE ASSIGNED AT DIFFERENT TIMES IN THE COURSE
(don't work on them until they are assigned)
The Theater Problem
1. Suppose three couples go to a theater and sit in adjoining seats. What is the probability that at least one couple is sitting side-by-side?
.
The Boy-Girl Paradox
2. Suppose a friend of yours is a married woman with two children, not twins. Order them by their age.
(i) Suppose she has one girl, what is the probablity both of her children are girls?
((ii) Suppose the youngest child is a girl, what is the probability both of her children are girls?
To do (i) and (ii), first draw the sample space without any conditions { BB, BG, GB, GG}, then draw the conditional sample spaces (they are different) and read off the answer for P(GG| one child is a girl) (i)
and P( GG| the first child is a girl) ( the answers are different)
Now things get tricky.
(iiiSuppose your friend tells you that one of her children is a girl.
What is the probability both of her children are girls? Assume boys and girls are equally likely.
Now you have to compute P(GG|your friend tells you that one of her children is a girl). The point of (iii) is that the event (your friend tells you that one of her children is a girl) is different
from (given one of the children is a girl) because she could have told you one of the children is a boy. For example P(your friend tells you that one of her children is a girl|BG) = 1/2.
To do this problem you have to use Bayes' Theorem so you have
P(GG|your friends tells you one of her children is a girl)
= (by Bayes' Theorem) [ P(your friend tells you one of her children is a girl|GG)] / [ P(your friend tells you one of her children is a girl|GG)P(GG)
+ P(your friend tells you one of her children is a girl|BG)P(BG) + P(your friend tells you one of her children is a girl|GB)P(GB) + P(your friend tells you one of her children is a girl|BB)P(BB) ]
For example P(your friend tells you one of her children is a girl|GB)= 1/2 and P(GB)= 1/4 so the third term in the denominator is 1/8.
(i),(ii) and (iii) have different answers.
>
There are some paradoxical features about this problem. See Wikipedia
"Boy or Girl Paradox" (though this Wikipedia article is not very clearly written).
The Monty Hall Problem (from ``Let's make a deal" a TV show in the fifties)
3. There are three closed doors. Behind one of them is a car (or something else you want) and behind each of the
other two is a goat(or something else you don't want). You pick a door but don't open it. Then the host Monty Hall opens another
door with a goat behind it. Monty then asks ``do you want to switch''? Should you switch or stick with your
original choice? Compute the probabilities of getting the car using the two different strategies ``stick'' or
``switch''.
The two strategies do not have the same success probability. This led to a big controversy in Parade
Magazine in the Washington Post about fifteen years ago. I was amazed that one statistics professor
(not from U of Md) wrote `` this is stupid, of
course they both have probability 1/2 ''. He was dead wrong, so don't believe everything your professors
(me included) tell you.
Extra Problem on the equality of the mean and the median
4. Let X be a continuous random variable. Suppose the graph of the density function f(x) has a point of symmetry
x
0 (this means that if you reflect the part of the curve to the right of the vertical line
through x
0 you get the part of the curve to the left of that vertical line).
(i) Prove that the mean and the median are both equal to x
0 so in particular they are equal.
(ii) Find an example of a continuous random variable X so that the graph of the density function
has no point of symmetry but the mean and the median of X are still equal.
THE COURSE LECTURES
In addition to the text, I want you to download my lectures.
HAND-PRINTED LECTURES
Here is a complete set of scanned
hand printed lectures (easy to read) available in pdf format.
Professor Millson's Handwritten
Lectures
SLIDES FOR MY COURSE LECTURES
Here are typed slides for my lectures - the slides I used in class.
MIDTERM AND FINAL DATES
The first midterm will be in the
lecture hall ARM 0126 on Thursday, October 18 and the second midterm will be in ARM0126 on
Thursday , November 15.
The final exam will be in ARM0126 on Wednesday, Dec. 12
from 8:00 to 10:00 AM.
The homework quizzes, exams and final will be conducted according to the following rules.
1. No notes and NO "CHEAT SHEET".
2. No text.
3. No laptops
4. No calculators
5. No cellphones.
I will provide you on the exam with the formulas/tables you will need.
HOMEWORK ARCHIVE
All previous homework assignments are archived here
HW archive -->
COURSE DESCRIPTION
We will cover the
following topics in the order given..
1. All of Chapter 2.
2. All of Chapter 3
3. All of Chapter 4 except for
the last two sections.
4. All of Chapter 5 but very little multiple
integrals.
5. Random Samples (from handout) .
6. Some of Chapter 6 on point estimation.
7. Random intervals (from handout).
8. Some of Chapter 7 on interval estimations.
9. Moment generating functions (from handout).
COURSE GRADE
Your course grade will be determined from the
following:
Test #1 50 points
Test #2 50 points
Homework 50 points
Final exam 100 points
Total 250
points
If you get 125 points or more you will get at worst a C-. If you get
less than 125 points you
will get an F.
Your homework grade.
The week's homework will be assigned on my web page after the Thursday class
THE HOMEWORK QUIZZES
For the first 35 minutes of your TA section on Monday your TA will do
the HW problems (if you are lucky he/she will do the one you are about
to be tested on). At the end (fifteen minutes left) of
your TA section your TA will pick one of
the HW problems, you will put away your notes, do the problem and hand it in.
In other words every week you will have a HW quiz. On occasion you may be
asked to hand in an entire assignment or a part of one.
If you have an excused absence from a HW quiz (TA session) you have two choices:
1. Do the entire assigment and hand it in
before the TA section in question (your TA will then grade the relevant question).
2. Otherwise you will receive the grade of E (excused) for the week's HW quiz and your homework total will be rescaled by the
appropriate factor (larger than one).
All HW quizzes will count (I will not discount your two worst quizzes as is sometimes done).
MAKEUPS
There will be no makeup for the
final exam or midterms except in case of medical emergency, family medical emergency, job interview
or university related absence. If you have a medical emergency we will
need a note verifying that from a medical professional. If you have a job interview we will need to see written evidence
of that interview and interview date.
In particular there will be no makeups
for family vacations so be careful about buying airplane tickets.
If you do miss a midterm or the final without being excused notify me that you have missed
the midterm or the final within a week. Otherwise you will receive a zero.
GRADING APPEALS
After each midterm or HW quiz students have one week from when the midterm
or quiz is returned to appeal the grading. No appeals for regrading work done during the semester
(including the midterms) can be made after the day of the final exam.
Appeals for the final exam must be done in writing.
BACKGROUND EXTRA COURSE NOTES (to be downloaded by you)
- To obtain a copy of the
handout on the basic probability distributions in pdf
format click distributions
- To obtain a copy of the lecture
Change of Continuous Random Variable in pdf
format click change of variable
-
The pdf
file for homework assignment (Extra Problem for HW 9) (Covariance and Correlation) is
available right here: ExtraProblemforHomework9
- The pdf
file for the lecture Random Samples is available right here: randomsamples
-
The pdf file for Practice Midterm 1 is available
right here: PracticeMidterm1
-
The pdf file for Practice Midterm 2 is not available here - go to Canvas for the link
-
The pdf file for the lecture on random intervals
is available right here: RandomIntervals
-
The pdf file for the lecture on confidence
intervals for the mean in a normal distribution when the variance is known
is available right here: ZConfidenceIntervals
-
The pdf file for the lecture on confidence
intervals for the mean in a normal distribution when the variance is
unknown is available right here: TConfidenceIntervals
-
The pdf file for the lecture on prediction
intervals for the next observation from a normal distribution when the
variance is known is available right here: ZPredictionIntervals
-
The pdf file for the lecture on prediction
intervals for the next observation from a normal distribution when the
variance is unknown is available right here: TPredictionIntervals
- Last Lecture. Moment Generating Functions
To obtain a copy of this lecture in pdf
format click momentgeneratingfunctions
-
The pdf file for the practice final exam is available
right here: PracticeFinal
The pdf file for the solutions of the practice final exam is available
right here: PracticeFinalSolutions