Random Variables and Probability Distributions

Recall

 Function - is a mapping from one set to another.

a. Domain – permissible values of input (sample points, outcome or elements)
b. Range – permissible values of output (value of x)
- function (f) relates the set of Reals to the set of Reals.

Random Variables - a function that maps each sample point to a real number.

 Events in terms of Random Variables

- 𝑋 ≤ 𝑎 - event containing all sample points whose associated value for the random variable X is
less than or equal to a
- 𝑋 > 𝑎 – event containing all sample points whose associated value for the random variable X is
greater than a
- 𝑎 < 𝑋 < 𝑏 - event containing all sample points whose associated value for the random variable
X is in between a and b

Types of Random Variables

 Discrete
- Defined over discrete sample space
- The sample space contains Countable sample points
- Possible values are Discrete points
- Uses pmf (easier) and cdf
- P(x) ≥ 0
- Sum is 1
 Continuous
- Defined over continuous sample space
- Uncountable
- Intervals
- Uses Pdf and cdf (easier)
- f(x) ≥ 0
- total area under the curve is 1

Probability Distributions

 Cumulative Distribution Function (F(●)) - is a function defined for any real number x as,
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥)
- We can use the probability mass function (PMF) to derive the CDF of the discrete random
variable.
- We can use the probability density function (PDF) to derive the CDF of the continuous random
variable. (INVOLVES INTEGRATION SO NOT INCLUDED)
 Discrete probability distribution - A table or formula listing all possible values that a discrete
random variable can take on, along with the associated probabilities
- Can be categorized into two: discrete cdf and pmf
 - The probabilities associated with all possible values of a discrete random variable must sum to
1.
a. Probability mass function (p(●)) – is a function defined for any real number x as,
𝑝(𝑥) = 𝑃(𝑋 = 𝑥)
- Mass points - values of x for which p(x) > 0
- The event that X=x is possible to happen
- usually presented in tabular form whenever X has only a few mass points.
- The sum of p(x) for all the mass points of X is 1.
 Continuous probability distribution
a. Probability distribution function (f (●)) - a function that is defined for any real number x and
satisfies the following:
1. f(x) ≥ 0 for all x
- values of a pdf are non-negative
2. the total area under its curve and above the horizontal axis is equal to 1
3. 𝑃(𝑎 ≤ 𝑋 ≤ 𝑏) - is the area bounded by the curve, the x-axis, and the lines x=a and x=b
b. Using CDF - Oftentimes, it is easier in computing for probabilities.
1. 𝑃(𝑋 ≤ 𝑎) = 𝑃(𝑋 < 𝑎) = 𝐹(𝑎)
2. 𝑃(𝑋 > 𝑎) = 𝑃(𝑋 ≥ 𝑎) = 1 − 𝐹(𝑎)
3. 𝑃(𝑎 < 𝑋 < 𝑏) = 𝑃(𝑎 ≤ 𝑋 ≤ 𝑏) = 𝑃(𝑎 < 𝑋 ≤ 𝑏) = 𝑃(𝑎 ≤ 𝑋 < 𝑏) = 𝐹(𝑏) − 𝐹(𝑎)
Expectation and Variance

 Expected Value of X (E(X)) of Discrete Variable

- Referred to as the mean of X
- FORMULA: 𝐸(𝑋) = 𝜇 = 𝑥1 ∗ 𝑝(𝑥1 ) + 𝑥2 ∗ 𝑝(𝑥2 ) + ⋯ + 𝑥𝑛 ∗ 𝑝(𝑥𝑛 ) = ∑𝑛𝑖=1 𝑥𝑖 ∗ 𝑝(𝑥𝑖 )
 Expected Value of g(X)
- the expected value of g(X) is
𝑛

𝐸(𝑔(𝑋)) = ∑ 𝑔(𝑥𝑖 ) ∗ 𝑝(𝑥𝑖 )

𝑖=1

 Variance
- Let X be the random variable with mean 𝜇, then the variance of X is

Definitional Formula

𝜎 2 = 𝑉𝑎𝑟(𝑋) = 𝐸[(𝑥 − 𝜇)2 ]

With the computational formula

𝑉𝑎𝑟(𝑋) = 𝐸(𝑋)2 − (𝐸(𝑋))2

- Easier to use if it has decimals
- 𝜇 𝑖𝑠 𝑒𝑞𝑢𝑎𝑙 𝑡𝑜 𝐸(𝑥)
1. Variance of X of Discrete Random Variable
- A measure of dispersion.
 𝑛
2 2]
𝜎 = 𝑉𝑎𝑟(𝑋) = 𝐸[(𝑥 − 𝜇) = ∑(𝑥𝑖 − 𝜇)2 ∗ 𝑝(𝑥𝑖 )
𝑖=1

 Standard Deviation of X
- the positive square root of the variance.
 Properties of the Mean – applicable for discrete and continuous
1. 𝐸(𝑋 − 𝜇) = 0
2. 𝐸(𝑎𝑋 + 𝑏) = 𝑎𝐸(𝑋) + 𝑏
- If 𝐸(𝑎𝑋) = 𝑎𝐸(𝑥) – a is a constant
- If 𝐸(𝑥 + 𝑏) = 𝐸(𝑥) + 𝑏 – b is a constant
3. 𝐸(𝑋 + 𝑌) = 𝐸(𝑋) + 𝐸(𝑌)
4. 𝐸(𝑋 − 𝑌) = 𝐸(𝑋) − 𝐸(𝑌)
 Properties of the Variance
1. 𝑉𝑎𝑟(𝑋) = 𝐸(𝑋 2 ) − [(𝐸(𝑋)]2
2. 𝑉𝑎𝑟(𝑎𝑋 + 𝑏) = 𝑎2 ∗ 𝑉𝑎𝑟(𝑋)
3. If X and Y are independent, then
3.a. 𝑉𝑎𝑟(𝑋 + 𝑌) = 𝑉𝑎𝑟(𝑋) + 𝑉𝑎𝑟(𝑌)
3.b. 𝑉𝑎𝑟(𝑋 − 𝑌) = 𝑉𝑎𝑟(𝑋) + 𝑉𝑎𝑟(𝑌)

Some Probability Distributions

 Parameter
- a constant that determines the specific form of the probability distribution.
- carries vital information like the shape of the distribution, the location of the distribution, and
other characterizations
 Binomial Distribution
- describes how frequent a “yes” appears in a “yes” / “no” experiment repeated for a fixed
number of time.
- Let X be the number of successes observed in n trials. From this, we can tell that X is a discrete
random variable. X is said to follow a binomial distribution if its pmf is given by:

𝑝(𝑥) = (𝑛𝑥 )𝑝 𝑥 (1 − 𝑝)𝑛−𝑥 , 𝑥 = 0,1,2, … , 𝑛

Where n (no. of trials) and p (probability that the outcome is a success) are parameters, P is any
value between 0 and 1, n is any positive integer

Simpler Formula
𝑋~𝐵𝑖(𝑛, 𝑝)
𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 − 𝐸(𝑋) = 𝑛𝑝
𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑜𝑓 𝑋 − 𝑉𝑎𝑟 (𝑋) = 𝑛𝑝(1 − 𝑝) = 𝑛𝑝𝑞
 1. Binomial Experiment - one that possesses the following properties:
a. consists of n identical trials.
b. Each trial results in one of two outcomes a “success” or a “failure” (called a Bernoulli
trial).
c. The probability of success on a single trial is equal to p (common probability of success)
and remains the same from trial to trial.
The probability of failure is equal to q = 1 – p
d. trials are independent.
- Binomial random variable – a random variable that counts the number of “successes” out of n
trials.
 Normal Distribution
- A continuous random variable X is said to be normally distributed or follows the normal
distribution if its PDF is
1 (𝑥−𝜇)2
−
𝑓(𝑥) = ℯ 2𝜎2
𝜎√2𝜋

If X follows the normal distribution, we could simply write this down as

𝑋~𝑁(𝜇, 𝜎 2 )

Where 𝐸(𝑋) = 𝜇 𝑎𝑛𝑑 𝑉𝑎𝑟(𝑋) = 𝜎 2

- graph of the normal distribution is called the normal curve/ bell curve.
- an arrangement of a dataset where most values cluster in the middle and the rest taper off
symmetrically towards either extreme.
- has two parameters – its mean and variance
 Properties of Normal Curve
1. normal curve is bell-shaped and symmetric about its mean 𝜇.
2. normal curve approaches the horizontal axis asymptotically as we proceed in either
direction away from the mean.
3. total area under the curve and above the x-axis is equal to 1.
 Standard Normal Distribution
- Useful in evaluation of probabilities and in inferential statistics
a. Standard Normal Random Variable (Z)
- the normal random variable has 𝜇 =0 and 𝜎 2 =1
- can be simply written as 𝑍~𝑁(0,1)
- To find the probability that Z is less than or equal to some value z, we can use what we call a
standard normal table or z-table.
- Z-table gives the cdf of the standard normal random variable at a specified value
Φ (𝑧) = 𝑃(𝑍 ≤ 𝑧)

- P(Z ≤ a) = P(Z < a) = F(a) = Φ(a)

- P(Z > a) = P(Z ≥ a) = 1 − F(a) = 1 − Φ(a)
- P(a < Z < b) = P(a < Z ≤ b) = P(a ≤ Z < b) = P(a ≤ Z ≤ b) = F(b) − F(a) = Φ(b) − Φ(a)
 b. Standard Score of a Normal Random Variable
- FORMULA FOR z-score

𝑋−𝜇
𝑍=
𝜎

- Whenever X is between the values 𝑥1 and 𝑥2 , the random variables Z will fall between the
corresponding values 𝑧1 and 𝑧2 . Thus,
𝑥1 − 𝜇 𝑋 − 𝜇 𝑥2 − 𝜇
P (𝑥1 < 𝑋 < 𝑥2 ) = 𝑃 ( < < ) = 𝑃(𝑧1 < 𝑍 < 𝑧2 ) = Φ(𝑧2 ) − Φ(𝑧1 )
𝜎 𝜎 𝜎
 CDF of a Normal Random Variable
- If we want to compute for probabilities of events involving a normal random va riable, then all
we need to do are
a. Transform X into a standard normal variable (z-score formula)
b. Use the Z table to evaluate the probability in terms of Z