These are the lecture notes to accompany STAT 3100.
There will be organized into a series of chapters (in two versions) that will be provided through UM Learn.
I expect you to know the mathematics in the pre-requisite courses for
STAT 3100.
Our goal in this course is to introduce tools and techniques that can be used to solve statistical problems such as constructing estimators, confidence intervals, and hypothesis tests.
4 / 17 (Some) Notations
Contents
1 Introduction
2 (Some) Notations
3 Sample spaces and Random Variables
4 Equal in Distribution, Random Variables
5 Reading
5 / 17 (Some) Notations
Sets
N = {1, 2, 3, . . .} : the set of natural numbers.
N0 = {0, 1, 2, 3, . . .} : the set of non-negative integers.
Z = {0, ±1, ±2, ±3, . . .} : the set of all integers.
Q = {a/b : a ∈ Z, b ∈ N} : the set of rational numbers (fractions).
R : the set of all real numbers.
6 / 17 (Some) Notations
Finite, Countably Infinite, and Uncountable Sets
A finite set has a finite number of elements.
A set S is said to be countably infinite if there exists a bijection from S to N. That is, there exists a function h from S to N such that for every element s ∈ S, h(s) = n ∈ N and h is invertible, so that for every n ∈ N, we have h−1 (n) = s ∈ S. A countably infinite set is ”enumerable” in the sense that we could create a (possibly infinite) list of all the possibilities. For example, N, N0 , Z, and Q are all countably infinite sets. A set that is either finite or countably infinite is simply said to be countable. An infinite set that is not countably infinite is said to be uncountable. For example, R is an uncountable set. The union of countable sets is still countable. Discrete random variables take values on countable sets. 7 / 17 Sample spaces and Random Variables
Contents
1 Introduction
2 (Some) Notations
3 Sample spaces and Random Variables
4 Equal in Distribution, Random Variables
5 Reading
8 / 17 Sample spaces and Random Variables
Sample space Definition 1 The sample space of a random experiment, which we will denote by Ω, is the set of all possible outcomes of that experiment.
Example 1 Suppose we roll two fair 6-sided die (one red and one blue). There are 36 possible outcomes, summarized as (R,B), where R is the number facing up on the red die and B is the number facing up on the blue die :
Random Variable Definition 2 A random variable, say X, defined on Ω, is a real-valued function from Ω to R. That is X : Ω → R (read ”X maps Ω into R”).
A probability model on Ω could reasonably assume that, on any given roll
of these dice, each of these outcomes is equally likely so that Pr (ω) = 1/36 for each ω ∈ Ω. We may define a number of random variables on this sample space : X is the sum of the two die ; Y is the maximum of the two die, etc. Each defines a function from the sample space to the real numbers. Example 2 ω (2,1) (4,3) (6,6) (3,2) (1,5) X (ω) = 3 7 12 5 6 Y (ω) = 2 4 6 3 5 10 / 17 Sample spaces and Random Variables
Probability of events Recall that an event is any subset of outcomes contained in the sample space Ω. Examples of events : A = {X = 3}, B = {Y ≤ 4}, etc. To calculate probabilities of events like A or B, we need to do so in terms of the probability structure defined on Ω. That is, we calculate Pr (X = 3)=Pr {ω ∈ Ω : X (ω) = 3} or Pr (Y ≤ 4)=Pr {ω ∈ Ω : Y (ω) ≤ 4}. We can use this concept to define probability distributions for X and Y. Example 3 x 2 3 4 5 6 7 8 9 10 11 12 Pr(X=x) 1/36 2/36 3/36 4/36 5/36 6/36 5/36 4/36 3/36 2/36 1/36 y 1 2 3 4 5 6 Pr(Y=y) 1/36 3/36 5/36 7/36 9/36 11/36
11 / 17 Sample spaces and Random Variables
Indicator function
Definition 3 Let E ⊆ Ω be an event. The indicator function of E , denoted by 1E , is the random variable defined as 1 if ω ∈ E 1E (ω) = 0 if ω < E
We sometimes write 1E (ω) = 1{ω ∈ E }.
We comment that we can use other subsets of R (not only events).
12 / 17 Equal in Distribution, Random Variables
Contents
1 Introduction
2 (Some) Notations
3 Sample spaces and Random Variables
4 Equal in Distribution, Random Variables
5 Reading
13 / 17 Equal in Distribution, Random Variables
Equal in distribution
Definition 4 Two random variables, say X and Y are said to be equal in distribution, or have the same distribution, if the cumulative distribution function (CDF) of X and Y are equivalent. That is, whenever FX (t) = FY (t) for all t at which FX (t) and FY (t) are both continuous.
Points of discontinuity can be ignored safely.
d The notation is X = Y . This does not require that X and Y are defined on the same sample space, only the CDF’s are equivalent. If X and Y are defined on the same sample space, this does not imply that X (ω) = Y (ω) for all ω ∈ Ω – only that X and Y have the same distribution.
14 / 17 Equal in Distribution, Random Variables
Example 4 Consider a Bernoulli experiment with events {X = 1} a success and {X = 0} a failure, and suppose that Pr {X = 1} = 1/2 (so that Pr {X = 0} = 1/2 also). Then Y = (X − 1)2 also has a Bernoulli distribution with p = 1/2. d However, X (ω) , Y (ω). We would say that X = Y , but not X = Y .
Example 5 Suppose that John has a fair penny (1 cent coin) and Jane has a fair quarter (25 cent coin). They each toss their coins 6 times. Let X be the number of heads that come up on John’s coin and let Y be the number of d heads that come up on Jane’s coin. Then X = Y sine both X and Y have a Binomial(6,1/2) distribution, but it is not necessarily true that X = Y for every set of outcomes.
15 / 17 Reading
Contents
1 Introduction
2 (Some) Notations
3 Sample spaces and Random Variables
4 Equal in Distribution, Random Variables
5 Reading
16 / 17 Reading
Reading
Review chapters 1-3 (Devore & Berk, 2018).
I am not expecting you to read all these 150 pages and know everything in them.
I just want you to refresh your memory on these topics.
