Denotation p(x) = P(X = x)
Probability density function 0 ≤ p(x) ≤ 1
(pdf) Characteristics
Interpretation of probability distribution
∑p(x) = 1
Denotation F(x) = P(X ≤ x)
Cumulative
distribution function
(cdf) Characteristics 0 ≤ F(x) ≤ 1
Uniform distribution: probability distributions
with equally likely outcomes
outcomes are discrete and have the same probability
Discrete uniform
distribution
pdf: p(1) = p(2) = p(3) = ... = p(n)
Characteristics
cdf: F(x) = P(X ≤ x)
Discrete random variable
two outcomes “failure” and “success”
Bernoulli trial
the trials are independent, with the
probability of success is constantly p
Binomial distribution
variable X where X is the number of successes in n Bernoulli trials
Definition of probability function: the probabilities
Binomial random variable Formula:
of the possible outcomes of a random variable
Binomial tree
Uniform distribution
defined over a range that spans between
Continuous uniform Outcomes can only occur between a and b
a (lower limit) and b (upper limit)
distribution
pdf:
R4: Common probability distribution
Characteristics
cdf:
Type something
Skewness = 0
Characteristics
Kurtosis = 3 and excess kurtosis = 0
mean = mode = median
probabilities of outcomes further above and below the
Random variable mean get smaller and smaller but do not go to zero
Univariate normal distribution probability of a single random variable
probability of a group of
related random variable
Multivariate normal distribution
n means of the n
Normal distribution series of returns
multivariate normal
distribution for the returns
on n assets n variances of the
n series of returns
0.5n(n – 1) pair-wise
correlations
Continuous random variable
Confidence intervals
normal distribution that has been
standardized so that it has a mean of 0
and a standard deviation of 1
Standard normal distribution
Formula
Application: using z-table
Type something
to determine probability
Shortfall risk
Assumption
Application: Roy’s safety first criterion
Formula:
Interpretation: The higher SFR, the more optimal a portfolio is
function:
skewed to the right
Charatceristics:
Lognormal distribution useful for modeling asset prices
bounded from below by zero
which never take negative values
Application: determine continuously compounded
annual rate for 2 consecutive periods:
fatter tail than normal distribution
Student’s t-distribution df = n – 1
the shape of t-distribution approaches normal distribution when df increases
asymmetrical (positively skewed)
Chi-square distribution
does not take on negative values
asymmetrical distributions bounded from below by 0
F distribution
the shape of F-distribution approaches normal distribution when both the numerator (df1)
and the denominator (df2) degrees of freedom increase
Monte Carlo simulation