Professional Documents
Culture Documents
Learning Objectives
To learn the concept of continuous random variable To compute probabilities from the normal distribution To use the normal probability plot to determine whether a set of data is approximately normally distributed To compute probabilities from the uniform distribution To compute probabilities from the exponential distribution To compute probabilities from the normal distribution to approximate probabilities from the binomial distribution
Probability Distributions
Probability Distributions Discrete Probability Distributions Binomial Poisson Hypergeometric
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
These can potentially take on any value, depending only on the ability to measure accurately.
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
f(X)
f(X) =
2 1 e (1/2)[(X)/] 2
Where
e = the mathematical constant approximated by 2.71828 = the mathematical constant approximated by 3.14159 = the population mean = the population standard deviation X = any value of the continuous variable
Empirical Rules
What can we say about the distribution of values around the mean? There are some general rules:
f(X) 1 encloses about 68% of Xs
-1
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
+1
68.26%
2 x
3 x
95.44%
99.73%
Values above the mean have positive Z-values, values below the mean have negative Z-values
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
Z=
f(Z) =
2 1 e (1/2)Z 2
Where
e = the mathematical constant approximated by 2.71828 = the mathematical constant approximated by 3.14159 Z = any value of the standardized normal distribution
Example
If X is distributed normally with mean of 100 and standard deviation of 50, the Z value for X = 200 is
Z=
This says that X = 200 is two standard deviations (2 increments of 50 units) above the mean of 100.
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
100 0
200 2.0
X Z
( = 100, = 50) ( = 0, = 1)
Note that the distribution is the same, only the scale has changed. We can express the problem in original units (X) or in standardized units (Z)
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
by the area
P (a X b) = P (a < X < b)
(Note that the probability of any individual value is zero)
a
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
0.5
0.5
0.9772
2.00
0.0 0.1
. . .
2.0
.9772
The value within the table gives the probability from Z = up to the desired Z value
8.6
Suppose X is normal with mean 8.0 and standard deviation 5.0. Find P(X < 8.6)
Z= X 8.6 8.0 = = 0.12 5.0
=8 = 10
=0 =1
8 8.6
0 0.12
.00
.01
.02
0.00 0.12
1.000
Z 0 0.12
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
Z 0 0.12
Z=
X 8 8 = =0 5
8 8.6 X Z 0 0.12
Z=
X 8.6 8 = = 0.12 5
.00
.01
.02
Z 0.12
X 8.0 7.4
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
0.0478
0.4522
X Z
Numerical Problems
Ref. # 5-38 Page No.270: Given that a random variable, X, has a normal distribution with mean 6.4 and standard deviation 2.7, find a. P(4.0 < X <5.0) b. P(X > 2.0) c. P(X < 7.2) d. P(X < 3.0) Ans. a. .1150 b. .9484 c. .6165 d. .8960
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
Numerical Problems
Ref. # 5-46 Page No.271: Glenn Howell, VP of personnel for the Standard Insurance Company, has developed a new training program, that is entirely self-paced. New employees work various stages at their own pace; completion occurs when the material is learned. Howells program has been especially effective in speeding up the raining process, as an employees salary during training is only 67% of that earned upon completion of the program. In the last several years, average completion time of the program was 44 days, and the standard deviation was 12 days.
a. b. c. What is the probability that an employee will finish the program in 33 to 42 days? What is the probability of finishing the program in fewer than 30 days? Fewer than 25 or more than 60 days?
a. b. c.
Numerical Problems
Ref. # 5-48 Page No.271: R. V. Poppin, the concession stand manager for the local hockey rink, just had 2 cancellations on his crew. This means that if more than 72000 people come to tonights hockey game, the lines for hot dogs will constitute a disgrace to Mr. Poppin and will harm business at future games. Mr. Poppin knows from experience that the number of people who come to the game is normally distributed with mean 67000 and standard deviation 4000 people. What is the probability that there will be more than 72000 people? Suppose Mr. Poppin can hire two temporary employees to make sure business wont be harmed in the future at an additional cost of $200. if be believes the future harm to business of having more than 72000 fans at the game would be $5000, should he hire the employees? Explain (Assume there will be no harm if 72000 or fewer fans show up, and that the harm due to too many fans doesnt depend on how many more than 72000 show up? Ans. 0.1056 Yes, the $200 cost is less than $528 expected loss to future business. (Since P(x>72000) = 0.1056 than expected loss $5000*.1056 = $528)
a. b.
a. b.
Numerical Problems
Ref. # 5-49 Page No.272: Maurine Lewis, an editor for a large publishing company, calculate that it requires 11 months on average to complete the publication process from manuscript to finished book, with a standard deviation of 2.4 months. She believes that the normal distribution well describes the distribution of publication times. Out of 19 books she will handle this year, approximately how many will complete the process in less than a year? Ans. 13 as P (x<12) = .6615
X = + Z
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
Example: Suppose X is normal with mean 8.0 and standard deviation 5.0. Now find the X value so that only 20% of all values are below this X
0.2000
? ?
8.0 0
X Z
Z -0.9
.03
.04
.05
? 8.0 -0.84 0
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
X Z
Numerical Problems
Ref. # 5-39 Page No.270: In a normal distribution with a standard deviation of 5.0, the probability that an observation selected at random exceeds 21 is 0.14 Find the mean of the distribution Find the value below which 4% of the values in the distribution lie. Find the value of Z such that P(X >21) = 0.14 or P(X < 21) = 0.86 The value of Z from the table +1.08 (+ sign because the area is on right of mu) = X- Z = 21- 1.08*5.0 = 15.6 X= 6.85 P (X< 6.85 ) = 4%
a. b. Ans.
a. b.
Numerical Problems
Ref. # 5-44 Page No.271: Jarrid Medical, Inc., is developing a compact kidney dialysis machine, but its chief engineer, Mike Crowe, is having trouble controlling the variability of the rate at which fluid moves through the device. Medical standards require that the hourly flow be 4 liter, plus or minus 0.1 liter, 80% pf the time. Mr. Crowe, in testing the prototype, has found that 68% of the time, the hourly flow is within 0.08 liter of 4.02 liters. Does the prototype satisfy the medical standards? Ans. Given P( 4.02-0.08 <X < 4.02+0.08) = 68% & obtain & then test the value of P(3.9 < X < 4.1) = 80%??? a. b. Since above P is 68% with in 4.02 0.8 implying = 4.02, = 0.8 P(3.9 < X < 4.1) = 0.7745 i.e. 77.45% not 80%
Numerical Problems
Ref. # 5-50 Page No.272: The Quickie Sales Corporation has just been given two conflicting estimates of sales for the upcoming quarter. Estimate I says that sales (in millions of dollars) will be normally distributed with = 325 and = 60. Estimate II says that sales will be normally distributed with = 300 and = 50. The board of directors finds that each estimated appears to be equally believable a priori. In order to determine which estimate should be used fro future predictions, the board of directors has decided to meet again at the end of the quarter to use updated sales information to make a statement about the credibility of each estimate. Assuming that estimate I is accurate, what is the probability that Quickie will have quarterly sales in excess of $350 million? Rework part (a) assuming that estimate II is correct. At the end of the quarter, the board of directors finds that Quickie Sales Corp. has had sales in excess of $350 million. Given this updated information, what is the probability that Estimate I was originally the accurate one? (Use Bayes Theorem) Rework part for Estimate II. Ans. 0.3385 0.1587 P(E1:x > 350) = ?
a. b. c.
d.
a. b. c.
Solution Ctd
P(E1 | D : X > 350) = = P(D | E1)P(E1) P(D | E1)P(E1) + P(D | E2)P(E2) (0.3385)(0.5) (0.3385)(0.5) + (0.1587)(0.5) 0.1692 = = 0.680 0.1692 + 0.0793
P(E2 | D : X > 350) = = P(D | E2)P(E2) P(D | E1)P(E1) + P(D | E2)P(E2)
Evaluating Normality
Not all continuous random variables are normally distributed It is important to evaluate how well the data set is approximated by a normal distribution
Evaluating Normality
(continued)
Assessing Normality
(continued)
A normal probability plot for data from a normal distribution will be approximately linear:
90 60 30 -2 -1 0 1 2
Left-Skewed
X 90
60 30 -2 -1 0 1 2 Z
Right-Skewed
X 90
60 30 -2 -1 0 1 2 Z
Rectangular
X 90
60 30
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
Numerical Problems
Example: The following ordered array (from left to right) depicts the amount of money (in dollars) withdrawn from a cash machine by 25 customers at a local bank: 40, 50, 50, 50, 50, 70, 70, 80, 80, 90,100, 100 90,100, 100 90,100, 100, 110, 110, 120, 120, 130, 140, 140, 150, 160, 160, 200 Decide whether or not the data appear to be approximately normally distributed by Evaluating the actual versus theoretical properties. Constructing a normal probability plot. Ans. Look at the MS-Excel worksheet
a.
300
200 Largest
200
100
80 Quartile 1 40 Smallest
0
N= 25
VAR001
Histogram
10
37.38 160
F requency
0.3849 0.2019
Std. Dev = 38.16 Mean = 106.8 N = 25.00 50.0 75.0 100.0 125.0 150.0 175.0 200.0
VAR001
Results
median =mode mean
1.00
(69,144) 18 (App. Equal) (59,155) 19 (App. Equal) (32,182) 24 (App. Equal) 49.7242 224.3197 50 (App. Equal) 160 (Not equal) .38 (App.
Equal)
Expected Cum Prob
.75
.50
.25
(continued)
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 General rule:
The normal distribution can be used to approximate the binomial distribution if np 5 and n(1 p) 5
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
(continued)
1 ba
if a X b
f(X) =
0 otherwise
where f(X) = value of the density function at any X value a = minimum value of X b = maximum value of X
2
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in
(b - a) 2 = 12
(6 - 2) 2 = 1 .1547 12
Example: Using the uniform probability distribution to find P(3 X 5): P(3 X 5) = (Base)(Height) = (2)(0.25) = 0.5
f(X) 0.25 X
2
Created by: Prabhat Mittal Email-Id: profmittal@yahoo.co.in