2022-09-15
Independence
[PSDR] Definition 2.20
Two events are said to be independent if knowledge that one event occurs does not give any probabilistic information as to whether the other event occurs. Formally, we say that \(A\) and \(B\) are independent if \(P(A \cap B) = P(A)P(B)\).
Events \(A\) and \(B\) are said to be dependent if they are not independent.
Conditional Probability
[PSDR] Definition 2.18
Let \(A\) and \(B\) be events in the sample space \(S\), with \(P(B) \ne 0\). The conditional probability of \(A\) given \(B\) is \[P(A|B) = \frac{P(A \cap B)}{P(B)}\]
Bayes’ Theorem and The Law of Total Probability
[PSDR] Definition 2.30
We say that \(A_1, \cdots, A_k\) is a partition of the sample space \(S\) if \(\bigcup_{i=1}^k A_i = S\) and \(A_i \cap A_j = \emptyset\) whenever \(i \ne j\).
[PSDR] Theorem 2.6
Let \(A_1, \cdots, A_k\) be a partition of the sample space \(S\) and let \(B\) be an event. Then,
\[P(B) = \sum_{i=i}^k P(B \cap A_i) = \sum_{i=1}^k P(A_i)P(B|A_i)\]
and
\[P(A_j|B) = \frac{P(A_j)P(B|A_j)}{\sum_{i=1}^k P(A_i)P(B|A_i)}.\]
In words,
\[\text{posterior} = \frac{\text{prior} \times \text{likelihood}}{\text{marginalization}}.\]
Consider a scenario where you have only one urn with contents one black ball and one red.
You draw one ball.
Sample space:
\[S = \{B,R\}\]
\(P(X = 1) = \frac{1}{2} \longrightarrow\) the probability of drawing black
\(P(X = 0) = \frac{1}{2} \longrightarrow\) the probability of drawing redWe can model the ball drawing using a probability function.
For one ball draw, we know the probability function to be \[P(X = x; p) = \begin{cases} p & \text{if } x=1 \text{ (B)} \\ 1-p & \text{if } x=0 \text{ (R)}\end{cases}\] where \(p=\frac{1}{2}\) and \(X\) is the discrete random variable for the outcome of the ball draw to be black or not black (which is red).
This function is also known as the Bernoulli Distribution, a statistical model for only two possible outcomes (yes/no,success/failure,etc.) of any single experiment.
We can rewrite the above function as \[P(X=x; p) = p^x (1-p)^{1-x}\]
The term \(p\) is the parameter of the function, which is the probability of observing a black ball.
You draw one ball with replacement three times. Note that each draw are now independent events.
Sample space: \[S = \{\{B,B,B\},\{B,B,R\},\{B,R,R\},\{R,R,R\}, \\ \{R,R,B\},\{R,B,B\},\{B,R,B\},\{R,B,R\}\}\]
Let the random variable \(X\) to be the number of black balls in this scenario; \[X = \{x_0,x_1,x_2,x_3\} = {0,1,2,3}\]
Example probabilities:
Here, \(S\) is a discrete sample space and \(X\) is a discrete random variable.
Let the discrete random variable \(X\) to be the number of black balls (say observing a black ball is considered as a success). We want to define a probability function to compute the probability of observing the number of black balls (observing the number of successes).
Let \(n\) be the number of tosses and \(p\) be the probability of success. In this example, \(p=\frac{1}{2}\).
The probability function that models multiple ball drawings from an urn - with replacement - is the Binomial, which is used when there are exactly two disjoint (also called mutually exclusive) outcomes of a trial (in this case, a ball draw from an urn with two distinct balls).
These outcomes are labeled “success” (observing heads) and “failure” (observing not heads - which tails).
The general form of the Binomial Probability Function is given by \[P(X = x; n, p) = \binom{n}{x} p^x (1-p)^{(n-x)} \hspace{5px} \text{ for } x = 0,1,2,\cdots, n\] where \(\binom{n}{x} = \frac{n!}{x!(n-x)!}\).
The parameters of this function is \(p\) and \(n\). The binomial distribution assumes that \(p\) is fixed for all \(n\) trials.
Notice that the term \(p^x (1-p)^{(n-x)}\) is the Bernoulli. Here, a trial is also called a Bernoulli trial, a random experiment with exactly two possible outcomes.
Using the binomial probability function, we can answer probability questions much more easily.
Using the binomial probability function, we can answer probability questions much more easily.
The probability of observing exacty 7 black balls is \[P(X = 7) = 0.0739\]
Using the binomial probability function, we can answer probability questions much more easily.
The probability of observing at least 7 black balls is \[P(X \ge 7) = \sum_{x=7}^{20} P(X = x) = 0.9423\]
Discrete Random Variable: A random variable that
can take on only a finite or countably infinite set of outcomes. We can
say that a discrete R.V. has distinct values that can be counted.
Example: Coin Flips, Dice Rolls, Outcomes of a Test
(positive/negative), gender (M/F), etc.
Continuous Random Variable: A random variable
that can take on any value along a continuum (but may be reported
“discretely”)
Example: Height (really continuous, but we
usually just report to the nearest inch/centimeter), temperatures,
etc.
Probability Functions = Probability Distributions Table, Graph, or Formula that describes values a random variable can take on, and its corresponding probability (discrete random variable) or density (continuous random variable).
We call a Probability Function as Probability Mass Function (PMF) for discrete random variables and Probability Density Function (PDF) for continuous random variables.
Mini-Assignment: Random Variables and Probability Functions
Back to Tentative Topics Schedule.