Professional Documents
Culture Documents
Chapter 4
Chapter 4
Chapter Four
Continuous Random Variables & Probability Distributions
Chapter Goals:
After completing this chapter, you should be able to
Determine probbilities from probability functions.
Determine probbilities from cumulative distribution functions, and the reverse.
Calculate the mean and the variance for continuous random variables.
Understand the assumptions for some common continuous probability ditributions.
Calculate the probabilities, determine means and variances for some common continuous
probability distributions
Standardize normal random variable.
Approximate probabiliteies for some binomial and Poisson ditributions
In general:
1. Every continuous random variable X has a curve associated with it.
2. This curve formally known as a probability density function (PDF), and it can be used to
obtain the probability associated with the random variable.
3. The cumulative distribution function (CDF) for a random variable X is given by
𝑥
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥) = ∫ 𝑓(𝑢) 𝑑𝑢, − ∞ < 𝑥 < +∞
−∞
Consider any two points a and b, where a less than b. Assume X is continuous random variable,
then
1. the probability that X between a and b is equal to the aria
under the curve
𝑃(𝑎 ≤ 𝑋 ≤ 𝑏) = 𝑡ℎ𝑒 𝑎𝑟𝑒𝑎 𝑢𝑛𝑑𝑒𝑟 𝑡ℎ𝑒 𝑐𝑢𝑟𝑣𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑎 𝑎𝑛𝑑 𝑏
𝑏
= ∫ 𝑓(𝑥) 𝑑𝑥 = 𝐹(𝑏) − 𝐹(𝑎)
𝑎
2. 𝑃(𝑋 = 𝑐) = 0
∞
3. The expexted value : 𝜇𝑥 = 𝐸(𝑋) = ∫−∞ 𝑥𝑓(𝑥) 𝑑𝑥
∞ ∞ 2
4. The variance :𝜎𝑥2 = 𝐸(𝑋 2 ) − 𝐸(𝑋)2 = 𝐸(𝑥 2 ) − 𝜇 2 = ∫−∞ 𝑥 2 𝑓(𝑥) 𝑑𝑥 − (∫−∞ 𝑥𝑓(𝑥) 𝑑𝑥)
P age |2
Example:
Suppose a random variable 𝑋 has a probability density function given by
1. Find the value of 𝑘 that makes this a probability density function (PDF)
2. Find 𝑃(0.4 ≤ 𝑥 ≤ 1)
3. Find 𝑃(𝑥 ≤ 0.4| 𝑥 ≤ 0.8)
4. Find the median.
5. Find the expected value and the variance of the random variable X.
6. For the random variable 𝑌 = 200𝑋 − 60, find the expected value and the variance
7. Find the cumulative distribution function (CDF) for a random variable.
P age |3
Example:
Specification call for the thickness of aluminum sheets that are to be made into cans be between 8
and 11 thousandths on an inch. Let 𝑓(𝑥) be the probability density function of 𝑋
𝑐 𝑥 6 ≤ 𝑥 ≤ 12
𝑓(𝑥) = {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Example:
The distribution function of the random variable X, the time (in months) from the diagnosis age
until death for one population of AIDS patients, is as follows:
1.3
𝐹 (𝑥 ) = 1 − 𝑒 −0.3𝑥 , 𝑥 > 0
1. Find the probability that a randomly selected person from this population survives at least
12 months.
2. Find the probability density function of X.
P age |5
Standard deviation:
12
Example:
A package delivery service divides their packages into weight classes. Suppose that packages in
the 14 to 20 pound class are uniformly distributed, meaning that all weights within that class are
equally likely to occur.
1. Find the probability that a package weights between 15 and 16.5 pounds.
2. Find the probability that a package weights less than 15 pounds.
3. Find the probability that a package weights at least 18 pounds.
4. Find the probability that a package weights exactly 15 pounds.
5. Find the minimum weight for the heaviest 20% of the packages.
6. Find the mean and the variance and the cumulative distribution for the random varialbe X
P age |6
An important (and not always desirable) characteristic of the exponential distribution is the so called lack
of memory property.
For 𝑥, 𝑠 > 0, consider the conditional probability 𝑃(𝑋 > 𝑥 + 𝑠 | 𝑋 > 𝑠) of the additional magnitude of an
exponentially distributed random variable given we already know it is greater than s.
Well
𝑃(𝑋 > 𝑥 + 𝑠 ⋂ 𝑋 > 𝑠) 𝑒 −𝜆(𝑥+𝑠)
𝑃(𝑋 > 𝑥 + 𝑠 | 𝑋 > 𝑠) = = = 𝑒 −𝜆𝑥
𝑃(𝑋 > 𝑠) 𝑒 −𝜆𝑠
So if we think of the exponential variable as the time to an event, then knowledge that we have waited time
s for the event tells us nothing about how much longer we will have to wait - the process has no memory.
P age |7
Example:
Life length of a particular type of battery follows exponential distribution with mean 200 hours. Find the
probability that the
1. life length of a particular battery of this type is less than 200 hours.
2. life length of a particular battery of this type is less than 200 hours or more than 400 hours.
3. life length of a particular battery of this type is more than 400 hours, given that it lives more that
300 hours.
P age |8
Example:
Customers arrive at the claims counter at the rate of 15 per hour. What is the probability that the arrival
time between consecutive customers is less than five minutes?
Time between arrivals is exponentially distributed with mean time between arrivals of 4 minutes
(15 per 60 minutes, on average)
1/ = 4.0, so = .25
P(x < 5) = 1 - e-a = 1 – e-(.25)(5) = .7135
Example
The probability density function of the time failure of an electronic component in a copier in
hours is
𝑓(𝑥) = 0.001𝑒 −0.001𝑥 , 𝑥 ≥ 0
1. Find the probability that a component lasts more than 3000 hours before failure.
2. Find the expected failure time.
3. Determine the number of hours at which 10% of all components have failed
P age |9
The Weibull distibution is good general purpose distribution , it may be used to represent a variety of
distributions.
- Used mostly because of its ability of fit data rather than its theoretical justification.
- Very common tool in reliability prediction.
The parameters in the distribution provide a great deal of flexibility to model systems in which
The number of failure increases with time (bearing wear)
Decreases with time (some semiconductors)
Remains constant with time (failures caused by external skocks to the system)
𝛽 𝑥 𝛽−1 −(𝑥)𝛽
𝑓(𝑥) = ( ) 𝑒 𝛿 𝑥>0
𝛿 𝛿
Note:
- When 𝛽 = 1, the Weibull distribution is
identical to the exponential distribution.
- The cumulative distribution function of the
random variable
𝑥 𝛽
−( )
𝐹(𝑥) = 1 −𝑒 𝛿
- The expected value and the variance for the Weibull distribution:
1
1. the mean : 𝜇 = 𝛿 Γ (1 + )
𝛽
2
2 1
2. the variance: : 𝜎 2 = 𝛿 2 Γ (1 + 𝛽) − 𝛿 2 (Γ (1 + 𝛽))
Definition:
𝑟
1. Γ(𝑟) = ∫0 𝑥 𝑟−1 𝑒 −𝑥 𝑑𝑥 𝑟 > 0
2. If 𝑟 integer, then Γ(𝑟) = (𝑟 − 1)!
3. For any 𝑟, Γ(𝑟 + 1) = 𝑟 Γ(𝑟)
1
4. Γ (2) = √𝜋
5. Γ(1) = 0! = 1
P a g e | 10
Example:
The time to failure (in hours) of a bearing in a mechanical shaft is modeled as a Weibull random variable
with β = ½ and δ = 5,000 hours.
Example:
2
The shear stringth in pounds of a spot weld is a Weibull distribution with 𝛽 = 3 and δ = 400 pounds
Find the following:
1. 𝑃(𝑋 > 410) =
2. 𝑃(𝑋 > 410| 𝑋 > 390) =
3. The mean and the variance
4. The 95th percentile
2
410 3
)
𝑃(𝑋 > 410) = 1 − 𝐹(410) = 𝑒 −(400
2
410 3
)
𝑃(𝑋 > 410 ∩ 𝑋 > 390) 𝑃(𝑋 > 410) 𝑒 −(400
𝑃(𝑋 > 410| 𝑋 > 390) = = = 2 = 0.96
𝑃( 𝑋 > 390) 𝑃( 𝑋 > 390) 390 3
−(400)
𝑒
1 1 5 3
𝐸(𝑋) = 𝜇 = 𝛿 Γ (1 + ) = (400) Γ (1 + ) = (400) Γ ( ) = (400) ( √𝜋)
𝛽 2 2 4
3
2
2 2
2 1 2 1 5
𝜎 2 = 𝛿 2 Γ (1 + ) − 𝛿 2 (Γ (1 + )) = 4002 Γ (1 + ) − 4002 (Γ (1 + )) = 4002 Γ(4) − 4002 (Γ ( ))
𝛽 𝛽 2 2 2
3 3
2
2 (3!) 2
3
= 400 − 400 ( √𝜋) = 357142.85
4
2 2
𝑎
)3 𝑎 3
𝑃(𝑋 < 𝑎) = 0.95 = 𝐹(𝑎) = 1 − 𝑒 −(400 → − ln 0.05 = ( ) → 𝑎 = 2074.02
400
P a g e | 11
Empirical Rules
What can we say about the distribution of values around the
mean? There are some general rules:
1. μ ± 1σ encloses about 68% of x’s
2. μ ± 2σ covers about 95% of x’s
3. μ ± 3σ covers about 99.7% of x’s
is a normal random variable with 𝐸(𝑥) = 0 𝑎𝑛𝑑 𝑉𝑎𝑟(𝑥) = 1. That is Z is a standard normal
random variable
Example:
Let Z has a standard Normal distribution, Find the following:
1. 𝑃(𝑍 < 1.26) =
2. 𝑃(𝑍 > −0.86) =
3. 𝑃(−.125 < 𝑍 < 0.3) =
4. Find a such that the P Z a 0.95
5. Find a such that the P Z a 0.1
6. Find a such that the P Z a 0.9
7. Find a such that the P Z a 0.8
8. Find Z 0.1
9. Find Z 0.05
P a g e | 13
Example:
The average life of a certain type of compressor is 10 years with standard deviation of 1 year. The
manufacturer replaces free all compressors that fail while under guarantee. Assume the lives of
the compressors follows a Normal distribution.
1. If a compressor selected randomly, find the probability that it has survived more than 11
years.
2. If they are willing to replace only 3% of all the compressors sold, how long a guarantee
should they offer?
P a g e | 15
Example:
The weekly amount spent for maintenance and repairs in a certain company has approximately a
normal distribution with a mena of $400 and a standard deviation of $20. If $450 is budgeted to
cover repairs for next week,
1. What is the probability that the actual costs will exceed the budgeted amount?
2. How much should be budgeted weekly for maintenance and repairs so that the budgeted
amount will be exceeded with probability only 0.1?
P a g e | 16
Continuity Correction
It is the adjustment made to an integer-valued discrete random variable when it is approximated by a
continuous random variable. For a binomial random variable, we inflate the events by adding or
subtracting 0.5 to the event as follows:
{X : X 4} {X : 3.5 X 4.5}
{X : X 4} {X : X 3}={X : X 3.5}
{X : X 4{ {X : X 4.5}
{X : X 4} {X : X 5} {X : X 4.5}
{X : X 4} {X : X 3.5} .
The continuity correction should be applied anytime a discrete random variable is being approximated by a
continuous random variable.
General rule:
The closer p is to 0.5, the better the normal approximation to the binomial
The larger the sample size n, the better the normal approximation to the binomial
The normal distribution can be used to approximate the binomial distribution if
np ≥ 5 and n(1 – p) ≥ 5
Example:
The manufacturing of semiconductor chips produces 2% defective chips. Assume the chips are
independent and that a lot contains 1000 ships.
1. Approximate the probability that more than 25 chips are defective.
2. Approximate the probability that between 20 and 30 chips are defective
P a g e | 18
Example:
Assume that the number of asbestos particles in a square meter of dust on a surface follows a
Poisson distribution with mean of 1000. If the squared meter of dust is analyzed, what is the
probability that 950 or fewer particles are found?
Example:
suppose that number of asbestos particles in a sample of one square centimeter of dust is a
Poisson random variable with a mean of 1000. Approximate the probability that ten square
centimeter of dust contains more than 10000 particles?
P a g e | 19
Example:
A high – volume printer produces minor print- quality errors on an a test pattern of 1000 pages of
text according to Poisson distribution with a mean of 0.4 per page.
1. What is the mean number of pages with errors (one or more)?
2. Approximate the probability that more than 350 pages contains errors( one or more)
P a g e | 20
1 (ln 𝑥−𝜃)2
−
𝑓(𝑥) = 𝑒 2𝜔2 , 0<𝑥<∞
𝑥𝜔√2𝜋
- The expected value and the standard deviation for the weibull
distribution:
𝜔2
1. the mean : 𝜇 = 𝑒 𝜃+ 2
2 2
2. the variance: : 𝜎 2 = 𝑒 2𝜃+𝜔 (𝑒 𝜔 − 1)
- The name follows from the transformation ln 𝑋 = 𝑊 . That is, the natural logarithm of 𝑋 is normally
distributed.
- Probabilities for 𝑋 are obtained from the transformation to 𝑊, but we need to recognize that the range
of 𝑋 is (0, ∞). Suppose that W is normally distributed with mean 𝜃 and variance 𝜔2 ; then the
cumulative distribution function for X is
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥)
= 𝑃(𝑒 𝑤 ≤ 𝑥) = 𝑃(𝑊 ≤ ln 𝑥)
𝑊 − 𝜃 ln 𝑥 − 𝜃
= 𝑃( ≤ )
𝜔 𝜔
ln 𝑥 − 𝜃
= 𝑃 (𝑍 ≤ )
𝜔
Example:
The lifetime of a semiconductor laser has a lognormal distribution with θ = 10 and ω = 1.5 hours.
1. What is the probability that the lifetime exceeds 10,000 hours?
2. What lifetime is exceeded by 99% of lasers?
3. What is the mean and variance of the lifetime?
Solution:
𝑃(𝑋 > 10000) = 1 − 𝑃(𝑒 𝑤 ≤ 10000)
= 1 − 𝑃(𝑊 ≤ ln 10000 )
𝑊 − 10 ln 10000 − 10
= 1−𝑃( ≤ )
1.5 1.5
= 1 − 𝑃(𝑍 ≤ −0.5264)
= 1 − 0.3 = 0.701
𝑃(𝑍 ≤ 𝑏) = 0.01
ln 𝑥 − 10
𝑏= = −2.33 → 𝑥 = 𝑒 6.505 = 668.48 ℎ𝑜𝑢𝑟𝑠
1.5 2 2
𝜔 1.5
𝜇 = 𝑒 𝜃+ 2 = 𝑒10+2 = 67846.29
P a g e | 21
2 2 2
𝜎 2 = 𝑒 2𝜃+𝜔 (𝑒 𝜔 − 1) = 𝑒 20+1.5 (𝑒1.5 − 1) = 39070059886.6 ℎ𝑜𝑢𝑟𝑠 2
Example:
Suppose that a mechanical component has wear – out effect that followis a lognormal with with θ = 8.5
and ω = 0.2 hours. What is the probability that the unit will last at least 3000 hours?