Probability Distribution

Chapter 4 4.1 Continuous Random Variables

Let X be a continuous rv. Then a
probability distribution or probability
Continuous Continuous Random A random variable X is continuous if its density function (pdf) of X is a function
set of possible values is an entire interval
Random Variables Variables and of numbers (If A < B, then any number x f (x) such that for any two numbers a
and Probability Probability Distributions between A and B is possible). and b, b
P ( a ≤ X ≤ b ) = f ( x)dx
a
Distributions The graph of f (x) is the density
curve.

Probability Density Function Probability for a Continuous rv

Probability Density Function Uniform Distribution
For f (x) to be a pdf
P(a ≤ X ≤ b) is given by the area of the shaded
If X is a continuous rv, then for any
1. f (x) > 0 for all values of x. A continuous rv X is said to have a
region. number c, P(x = c) = 0. For any two
2.The area of the region between the graph uniform distribution on the interval [A, B] numbers a and b with a < b,
of f and the x – axis is equal to 1. if the pdf of X is
y = f ( x) 1 P ( a ≤ X ≤ b) = P ( a < X ≤ b)
A≤ x≤ B
f ( x; A, B ) = B − A = P ( a ≤ X < b)
y = f ( x)
0 otherwise
Area = 1 = P ( a < X < b)
a b

The Cumulative Distribution Function Using F(x) to Compute Probabilities

Obtaining f(x) from F(x)
4.2 The cumulative distribution function, F
Let X be a continuous rv with pdf f(x) If X is a continuous rv with pdf f(x)
(x) for a continuous rv X is defined for
and cdf F(x). Then for any number a, and cdf F(x), then at every number x
Cumulative Distribution every number x by
x P ( X > a ) = 1 − F (a) for which the derivative F ′( x) exists,
Functions and Expected F ( x) = P ( X ≤ x ) = f ( y )dy F ′( x) = f ( x).
−∞
and for any numbers a and b with a < b,
Values For each x, F(x) is the area under the
density curve to the left of x. P ( a ≤ X ≤ b ) = F (b) − F (a )

Percentiles Median Expected Value Expected Value of h(X)

Let p be a number between 0 and 1. The The median of a continuous distribution, The expected or mean value of a If X is a continuous rv with pdf f(x) and h
(100p)th percentile of the distribution of a denoted by , is the 50th percentile. So continuous rv X with pdf f (x) is (x) is any function of X, then
continuous rv X denoted by η ( p ), is satisfies 0.5 = F( ) That is, half the area ∞ ∞
defined by under the density curve is to the left of µX = E ( X ) = x ⋅ f ( x) dx E [ h( x) ] = µh ( X ) = h( x) ⋅ f ( x) dx
η ( p) −∞ −∞
p = F (η ( p ) ) = f ( y )dy
−∞
 Variance and Standard Deviation Exponential Distribution Mean and Variance
Short-cut Formula for Variance
A continuous rv X has an exponential
The variance of continuous rv X with pdf The mean and variance of a random
distribution with parameter λ if the pdf is
f(x) and mean is µ variable X having the exponential
∞ ( )
V ( X ) = E X 2 − [ E ( X )]
2
distribution
σ X2 = V ( x) = 2
( x − µ ) ⋅ f ( x) dx λ e−λ x x≥0
f ( x; λ ) =
−∞ 0 otherwise 1 1
= E[( X − µ ) ]
2 µ = αβ = σ 2 = αβ 2 =
λ λ2
The standard deviation is σ X = V ( x).

Cdf of exponential Distribution Applications of the Exponential Normal Distributions

and its memoryless property Distribution
Suppose that the number of events 4.3 A continuous rv X is said to have a
Let X have a exponential distribution occurring in any time interval of length t normal distribution with parameters
Then the cdf of X is given by has a Poisson distribution with parameter α t The Normal µ and σ , where − ∞ < µ < ∞ and
and that the numbers of occurrences in 0 < σ , if the pdf of X is
0 x<0 nonoverlapping intervals are independent Distribution
F ( x; λ) = of one another. Then the distribution of 1
1 − e −λx x ≥ 0
2 /(2σ 2 )
f (x ) = e −( x −µ ) −∞ < x < ∞
elapsed time between the occurrences of σ 2π
two successive events is exponential with
parameter λ = α .

Standard Normal Distributions Standard Normal Cumulative Areas c. P(−2.1 ≤ Z ≤ 1.78)

Standard Normal Distribution
The normal distribution with parameter Find the area to the left of 1.78 then
Let Z be the standard normal variable.
values µ = 0 and σ = 1 is called a Shaded area = Φ (z ) subtract the area to the left of –2.1.
Standard Find (from table)
standard normal distribution. The normal
random variable is denoted by Z. The pdf a. P( Z ≤ 0.85) = P( Z ≤ 1.78) − P( Z ≤ −2.1)
curve
is 1 2 = 0.9625 – 0.0179
f ( z ;0,1) = e−z / 2 − ∞ < z < ∞ Area to the left of 0.85 = 0.8023
σ 2π 0 z = 0.9446
The cdf is b. P(Z > 1.32)
z
Φ ( z ) = P( Z ≤ z ) = f ( y;0,1) dy 1 − P( Z ≤ 1.32) = 0.0934
−∞

Ex. Let Z be the standard normal variable. Find z if a.

zα Notation P(Z < z) = 0.9278.
Normal Curve
Nonstandard Normal Distributions
Look at the table and find an entry Approximate percentage of area within
zα will denote the value on the = 0.9278 then read back to find
If X has a normal distribution with mean given standard deviations (empirical
measurement axis for which the area z = 1.46. µ and standard deviation σ , then rule).
99.7%
under the z curve lies to the right of zα .
b. P(–z < Z < z) = 0.8132 X −µ 95%
Shaded area P(z < Z < –z ) = 2P(0 < Z < z)
Z= 68%
= P( Z ≥ zα ) = α
σ
= 2[P(z < Z ) – ½]
= 2P(z < Z ) – 1 = 0.8132 has a standard normal distribution.
P(z < Z ) = 0.9066
0 zα z = 1.32
Ex. Let X be a normal random variable Ex. A particular rash shown up at an
with µ = 80 and σ = 20. 3.75 − 6 9− 6 Percentiles of an Arbitrary Normal
elementary school. It has been determined P ( 3.75 ≤ X ≤ 9 ) = P ≤Z ≤
Find P( X ≤ 65). that the length of time that the rash will 1.5 1.5 Distribution
last is normally distributed with = P ( −1.5 ≤ Z ≤ 2 )
65 − 80 µ = 6 days and σ = 1.5 days.
P ( X ≤ 65 ) = P Z ≤ (100p)th percentile (100 p )th for
20 for normal ( µ , σ ) = µ + standard normal ⋅ σ
Find the probability that for a student = 0.9772 – 0.0668
= P ( Z ≤ −.75 ) selected at random, the rash will last for
between 3.75 and 9 days. = 0.9104
= 0.2266

Normal Approximation to the Ex. At a particular small college the pass rate
Binomial Distribution of Intermediate Algebra is 72%. If 500
students enroll in a semester determine the
Let X be a binomial rv based on n trials, each probability that at least 375 students pass.
with probability of success p. If the binomial µ = np = 500(.72) = 360
probability histogram is not too skewed, X may
be approximated by a normal distribution with σ = npq = 500(.72)(.28) ≈ 10
µ = np and σ = npq .
375.5 − 360
P( X ≤ 375) ≈ Φ = Φ (1.55)
x + 0.5 − np 10
P( X ≤ x ) ≈ Φ
npq = 0.9394