Continuous Probability Distributions

Here will discuss several parametric families of univariate probability density functions.
Uniform Distribution
It is a family of symmetric probability distribution such that for each member of the family,
all intervals of the same length on the distribution’s support are equally probable.
1
𝑓𝑋 (𝑥; 𝑎, 𝑏) = {𝑏 − 𝑎 , 𝑎<𝑥<𝑏
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

Parameters 𝑎 and 𝑏 satisfies −∞ < 𝑎 < 𝑏 < ∞, then random variable 𝑋 is defined to be uniformly
distributed over the interval [𝑎, 𝑏].
𝑎+𝑏 (𝑏−𝑎)2
𝐸(𝑋) = 𝑉(𝑋) =
2 12

Example:
When a motorist stops at a red light at a certain intersection, the waiting time for the light to turn
green, in seconds is Uniformly distributed on the interval (0,30). Find the probability that the
waiting time is between 10 and 15 seconds.
Let X – Waiting time and 𝑋~𝑈(0,30).
1
, 0 < 𝑥 < 30
Therefore, probability density function 𝑓(𝑥) = {30
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
15
1 1 1
𝑃(10 < 𝑋 < 15) = ∫ 𝑑𝑥 = [𝑥]15
10 =
30 30 6
10

Exponential Distribution
Exponential distribution is used to model the waiting time to an event in a Poisson process.
−𝜆𝑥
𝑓𝑋 (𝑥; 𝜆) = {𝜆𝑒 , 0<𝑥<∞
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
λ is a parameter where 𝜆 > 0. It is called as rate parameter.
1 1
𝐸(𝑋) = 𝜆 𝑉(𝑋) = 𝜆2

Example:
The lifetime of a particular integrated circuit has an exponential distribution with mean 2 years.
Find the probability that the circuit lasts longer than 3 years.
Let T- the lifetime of a circuit
1
𝜇 = 2 and 𝜆 = 2 = 0.5

𝑃(𝑇 > 3) = 1 − 𝑃(𝑇 ≤ 3) = 𝑒 −0.5×3 = 0.223

Assume that the circuit is now 4 years old and is still functioning. Find the probability that it
functions for more than 3 additional years.

𝑃((𝑇 > 7) 𝑎𝑛𝑑 (𝑇 > 4))

𝑃(𝑇 > (4 + 3)|𝑇 > 4) =
𝑃(𝑇 > 4)
𝑃(𝑇>7) 𝑒 −0.5(7)
= 𝑃(𝑇>4) = 𝑒 −0.5(4) = 𝑒 −0.5(3) = 0.223

• Memoryless Property
𝑃(𝑇 > 𝑠 + 𝑡|𝑇 > 𝑠) = 𝑃(𝑇 > 𝑡)
If 𝑇 is conditioned on a failure to observe the event over some initial period of time 𝑠, the
distribution of the remaining waiting time is the same as the original unconditional distribution.
Gamma Distribution
Gamma distribution is used to extend the usefulness of the exponential distribution in
modeling waiting time.
Waiting time until r (positive integer) events have occurred is distributed as 𝜏(𝑟, 𝜆).
 𝜆𝑟 𝑟−1 −𝜆𝑥
𝑥 𝑒 , 𝑥>0
𝑓𝑋 (𝑥; 𝑟, 𝜆) = {𝜏(𝑟)
0, 𝑥≤0
𝑟 > 0, 𝜆 > 0 are parameters and 𝜏(𝑟) is a gamma function.
∞
𝜏(𝑟) = ∫0 𝑡 𝑟−1 𝑒 −𝑡 𝑑𝑡 𝜏(𝑟) = (𝑟 − 1)!

𝜏(𝑟 + 1) = 𝑟𝜏(𝑟)

𝜏(1⁄2) = √𝜋

When 𝑟 = 1 gamma distribution is same as the exponential distribution.

If 𝑋1 , 𝑋2 , … , 𝑋𝑟 are independent random variable each distributed as 𝐸𝑥𝑝(𝜆), then the sum 𝑋1 +
𝑋2 + ⋯ + 𝑋𝑟 is distributed as 𝜏(𝑟, 𝜆).
𝑟 𝑟
𝐸(𝑋) = 𝜆 𝑉(𝑋) = 𝜆2

Normal Distribution
In probability theory, the normal (Gaussian) distribution is a very common continuous
probability distribution. Normal distributions are often used in the natural and social sciences to
represent real-valued random variables whose distributions are not known.
1 −
1
(𝑥−𝜇)2
𝑓𝑋 (𝑥; 𝜇, 𝜎) = 𝑒 2𝜎2 −∞<𝑥 <∞
√2𝜋𝜎
Where parameters 𝜇 and 𝜎 satisfies −∞ < 𝜇 < ∞ and 𝜎 > 0.

If a random variable 𝑋 is normally distributed with mean 𝜇 and variance 𝜎 2 then we will write
𝑋~𝑁(𝜇, 𝜎 2 )
𝐸(𝑋) = 𝜇 𝑉(𝑋) = 𝜎 2
Standardizing a variable that has any normal distribution produces a new variable that has the
standard normal distribution.
When random variable 𝑋~𝑁(0,1) then it becomes standard normal distribution.
𝑋−𝜇
If 𝑋~𝑁(𝜇, 𝜎 2 ), then standardized variable 𝑍 = has the normal distribution.
𝜎

Example:
1) Test on two types of electric bulb shows the followings.
➢ Type A – Lifetime distributed normally with an average life of 1150 hours
and a standard deviation of 30 hours
➢ Type B – Lifetime distributed normally with an average life of 1900 hours
and a standard deviation of 50 hours
a) What percentage of bulbs of type A could be expected to have a life of more than
1200 hours?
b) What percentage of type B would you expect to last longer than 1800 hours?
c) What lifetime limits would you estimate would contain the central 80% of the
production of type A?
2) The speed of car passing a certain point on a motorway can be taken to be normally
distributed. Observations show that of cars passing the point, 95% are traveling at less than
85m.p.h and 10% are travelling at less than 55m.p.h.
a) Find the average speed of the cars passing the point
b) Find the proportion of cars that travel at more than 70m.p.h.

Normal Approximation to the Binomial Distribution and Poisson Distribution

Distribution of X Restriction on Parameters Approximation
𝑋~𝐵𝑖𝑛(𝑛, 𝑝) 𝑛 large and 𝑝 close to 0.5 𝑋~𝑁(𝑛𝑝, 𝑛𝑝𝑞)
OR
𝑛𝑝 > 5; 𝑛(1 − 𝑝) > 5
𝑋~𝑃𝑜𝑖(𝜆) 𝜆 > 20 𝑋~𝑁(𝜆, 𝜆)

Continuity Correction
Continuity correction is an adjustment that is made when a discrete distribution is approximated
by a continuous distribution. To do the continuity correction adding or subtracting 0.5 to a discrete
𝑋 value.

𝑃(𝑋 ≤ 𝑥) = 𝑃(𝑋 ≤ 𝑥 + 1) ≈ 𝑃(𝑌 ≤ 𝑥 + 1⁄2)

Example:
1) Find the probability of obtaining between 4 and 7 heads inclusive with 12 tosses of a fair
coin, using a Normal approximation to the Binomial distribution.
2) A radioactive disintegration gives counts that follows a Poisson distribution with mean
count per second of 25. Find the probability that in 1 second the count is between 23 and
27 exclusives, using the Normal approximation to the Poisson distribution.