Econometrics I
Lecture 3: Probability Refresher
Dr. Ernesto M. Gavassa-Pérez
egavass@unav.es
University of Navarra
Department of Economics
Spring, 2024
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 1 / 17
Roadmap
1 Terminology
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 2 / 17
Roadmap
1 Terminology
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 3 / 17
Probability
Terminology
A set is a group of elements together, and is denoted by a capital
letter
X := {1, 2, 3}
The previous set contains elements 1, 2, and 3
The sample space is denoted by S and denotes all possible outcomes
S = {W in, Lose}
= {Rain, Sunny}
= {Heads, T ails}
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 4 / 17
Probability
Terminology
What is a random variable?
You don’t know what will happen
Each element in S has a probability
e.g., P r(W in) = 0.4 & P r(Lose) = 0.6
More formally, f : S 7→ P is its distribution
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 5 / 17
Probability
Terminology
What is an event?
A collection of outcomes
Important: Don’t confuse it with outcome
An event can include several outcomes
P r(dice = 2) 7→ Simple Event
P r(dice = 2 or 4 or 6) 7→ Complex event
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 6 / 17
Probability
Probability Mass, Density & Cummulative distribution Functions: Intuition
Figure 1: pdf and cdf of a function
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 7 / 17
Probability
Probability Mass & Density Functions: Properties
Discrete Random Variables: PMF – Properties
X
f (X = xi ) ≥ 0 ∀xi ∈ S, and f (X = xi ) = 1
xi ∈X
Continuous Random Variables: PDF – Properties
Z ∞
f (xi ) ≥ 0 ∀xi ∈ S, and f (xi )dxi = 1
−∞
The probability of each outcome has to be non-negative
The sum of probabilities of all outcomes = 1
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 8 / 17
Probability
Probability Mass Functions: Examples
Rolling a die: Probability Mass Function
0
x≤0
1
P r (X = x) = x ∈ {1, 2, 3, 4, 5, 6}
6
0 x≥6
Figure 2: pmf of rolling a fair dice
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 9 / 17
Probability
Probability Mass Functions: Examples
Slightly more complex: Tossing two fair coins
Step 1. Define Sample Space
S := {(H, H), (H, T ), (T, H), (T, T )}
Step 2. Define probabilities of each possible event
▶ P r(X = Two heads) = 1
4
▶ P r(X = Two Tails) = 1
4
▶ P r(X = one head & one tail) = 1
2
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 10 / 17
Probability
Probability Density Function: Examples
f(x) = 3 ·x2
Z 0.4
P r (0.2 ≤ X = x ≤ 0.4) = 3 · x2 dx
0.2
3
1
0.8
2
0.6
f(x)
F(x)
1 0.4
0.2
0
0 0.2 0.4 0.6 0.8 1 0
0 0.2 0.4 0.6 0.8 1
xx
Interval xx
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 11 / 17
Probability
Cummulative Distribution Functions
CMF – Properties
P r (X ≤ xi ) ≥ 0 ∀xi ∈ S and P r X ≤ sup =1
xi ∈S
Prob. of X being lower than any given outcome has to be > 0
The probability of being lower than the highest outcome = 1
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 12 / 17
Probability
Cummulative Distribution Function
CMF – Discrete Random Variables
k
X
P r (X ≤ xk ) = P r (X = xi )
i=−∞
CMF – Continuous Random Variables
Z xi
P r (X ≤ xi ) = f (xi )dxi ∀xi ∈ S
−∞
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 13 / 17
Probability
Cummulative Distribution Functions: Examples
Rolling a die: Cummulative Distribution Function
0
x<1
k 1
P r (X ≤ x) =
P
i=1 6 x≤k
1 x≥6
Figure 3: pmf & cdf of rolling a fair dice
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 14 / 17
Probability
Cummulative Distribution Function: Examples
Rolling a die: Cummulative Distribution Function
1
P r(X ≤ 1) = 6
P4 1 1 4
P r(X ≤ 4) = i=1 6 =4· 6 = 6
etc.
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 15 / 17
Probability
Cummulative Distribution Functions: Examples
A continuous example
Assume X ∈ [0, 1]
Z x
P r(X ≤ x) = 3 · x2 dx
0
R 0.2 0.2
P r(X ≤ 0.2) = 0 3 · x2 dx = x3 0 = 0.008
R 0.5 0.5
P r(X ≤ 0.5) = 0 3 · x2 dx = x3 0 = 0.125
R 0.9 0.9
P r(X ≤ 0.9) = 0 3 · x2 dx = x3 0 = 0.729
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 16 / 17
Probability
Exercises I like
Find value of c so that they are valid pdf’s
f (x) = cx2 ; x ∈ [0, 2]
Z 2
c · x2 dx = 1
0
c 3 2
· x 0=1
3
8
c· =1
3
3
c=
8
Ernesto M. Gavassa-Pérez (Unav) Econometrics I Spring, 2024 17 / 17
