Professional Documents
Culture Documents
Terminology
Random Variable
Probability Distributions
The value of a random variable can not be predicted accurately, by convention, the probabilities are assigned to all the likely values that the variable may take. By enumerating the possible values & assigning probabilities specifically to each of these values, we are in fact, describing a probability distribution. Expected Value E(X) = [(X) * P(X)]
Example: Question
Q. A person expects a gain of Rs. 80, Rs. 120, Rs.160 & Rs. 20 by investing in a share. The probability distribution of the gain is as follows:
Gain (Rs.) 80 120 160 20 Probability 0.2 0.4 0.3 0.1
The expected gain from the share is = (80*0.2)+(120*0.4) + (160*0.3)+(20*0.1) = Rs. (16+48+48+2) = Rs.114
Example 2
Q. An Investor is contemplating to invest either in bonds that will pay 9% interest or in stocks. The stock prices may either go up by 20% or decrease by 10% or remain unchanged. The probabilities of these three occurrences were found by the investor as 0.3, 0.2 & 0.5 respectively. If the investor has initial capital of Rs. 1000, what should he do?
Solution 2
Expected Money Value from Bonds: Interest from Bonds = 1000*9/100 = 90 Total EMV = 1090
Price of X 1200 900 1000 Total E(X) P (X) 0.3 0.2 0.5 E(X) 360 180 500 1040
Comparing the EMV of both the options, Bonds EMV is higher so it is advisable to invest in Bonds
Variance:
The variance is a measure of the variability or dispersion in a variable or a data set. Variance = (S.D.)2
Covariance
A measure of variability of one variable (or a data set) in relation to another variable (or a data set), is the covariance.
Continuous Probability
The variables are allowed to take on any value within a given range. Example:
Normal Distribution; so, we cannot list all the possible outcomes.
Binomial Distribution
It describes discrete, non-continuous data resulting from an experiment that is also known as Bernoulli process. P(X=r) = nCr * pr*qn-r = n!/ (n-r)!* (r!)* pr*qn-r It has an expected value (or Mean ) = np where, n = total no. of trials;
p = probability of success
Variance = 2 = npq where, q = probability of failure
Solution:
P = 4C3 * ()3 * ()1 = (4*27*1)/ 64*4 = 27/64 = 0.42
(X=3)
Example 2
Q. If 5% of the items produced turn out to be defective, then find out the probability that out of 20 items selected at random there are: i. Exactly three are defective ii. Atleast two are defective iii. Exactly four are defective iv. Find the mean & variance
i.
Note: is the only parameter [tell me and I can calculate the probabilities] and e is the natural logarithm base.
FYI:
7.17
Poisson Distribution
As mentioned on the Poisson experiment slide: The probability of a success is proportional to the size of the interval Thus, knowing an error rate of 1.5 typos per 100 pages, we can determine a mean value for a 400 page book as:
Example 7.13
For a 400 page book, what is the probability that there are no typos? e=2.7183 P(X=0) =
there is a very small chance there are no typos
7.19
Example 7.13
For a 400 page book, what is the probability that there are five or less typos? P(X5) = P(0) + P(1) + + P(5) This is rather tedious to solve manually. k=5, =6, and P(X k) = .446
Example
The number of typographical errors in new editions of textbooks varies considerably from book to book. After some analysis he concludes that the number of errors is Poisson distributed with a mean of 1.5 typos per 100 pages. The instructor randomly selects 100 pages of a new book. What is the probability that there are no typos?
= 1.5?
7.21
Poisson Distribution
It is used to approximate binomial distribution when: n is large & p is very small. Conditions: Occurrence of each event is independent of the other. P (X=x) = [(e-m)*(mx )] / x! Where, m = mean x = no. of occurrences of the random event e = constant Mean = m = variance
Normal Distribution
Normal distribution reflects the various values taken by variables like heights/ weights of people. A large no. of observations are found to be clustered around the mean values & their frequency drops sharply as we move away from the mean in either direction.
Normal Distribution
When n is large, n tends to infinity and p is not that small, i.e., p,q tends to equality, we use this method Standard Normal Variate (Z) Z = X np / npq
Normal Distribution
68-95-99.7 Rule
68-95-99.7 Rule
Example
For example: Whats the probability of getting a math SAT score of 575 or less, =500 and =50?
575 500 Z 1.5 50 i.e., A score of 575 is 1.5 standard deviations above the mean
But to look up Z= 1.5 in standard normal chart (or enter into SAS) no problem! = .9332
Practice problem
If birth weights in a population are normally distributed with a mean of 109 oz and a standard deviation of 13 oz, a. What is the chance of obtaining a birth weight of 141 oz or heavier when sampling birth records at random? b. What is the chance of obtaining a birth weight of 120 or lighter?
Answer
a. What is the chance of obtaining a birth weight of 141 oz or heavier when sampling birth records at random?
141 109 Z 2.46 13
From the chart or SAS Z of 2.46 corresponds to a right tail (greater than) area of: P(Z2.46) = 1-(.9931)= .0069 or .69 %
Answer
b. What is the chance of obtaining a birth weight of 120 or lighter?
120 109 Z .85 13
From the chart or SAS Z of .85 corresponds to a left tail area of: P(Z.85) = .8023= 80.23%
Solution:
Z = (X- ) / X=0.2 = 10, = 5 Z = (0.2-10)/ 5 = -1.96 Z = (X- ) / X = 19.8 = 10, = 5 Z = (19.8-10)/ 5 = 1.96 Area under the curve from -1.96 to 1.96 0.475*2 = 0.95
Q2. Suppose that 4% of all TVs made by W&B Company in 1995 are defective. If eight of these TVs are randomly selected from across the country and tested, what is the probability that exactly three of them are defective? Assume that each TV is made independently of the others.
Q3. Take S = {1, 2, 3, 4, 5, 6}. The following table gives a probability distribution for S.
Outcome Probability 1 .3 2 .3 3 0 4 .1 5 .2 6 .1
Sol:
Q2.: n=8, X=3, p=0.04, and q=(1-p)=0.96 P(X) = P(3) = 0.0003 or 0.03% Mean = (n) x (p) =(8)(0.04)=0.32 Variance = np (1 - p) = (0.32)(0.96) = 0.31 Standard deviation = 0.31 = 0.6.
Q4. In a survey of voters, 30% of the respondents reported supporting the Republican candidate in the last presidential election, and 35% reported supporting the Democrat candidate. (Voting machines made it impossible to "spoil the ballot" by voting for both candidates.) Based on this data, the experimental probability that a voter will vote for a presidential candidate of either major political party is:
Q5. Binomial Random Variable: A binomial random variable is a random variable that counts the number of successes in a sequence of independent Bernoulli trials with fixed probability of success.
Now decide which of the following is a binomial random variable. a) Throw two dice 10 times; X = the number of doubles. X is
b) A bag of marbles contains 10 red ones and 10 blue ones. c) You draw 10 of them; X = the number of red ones. X is
Q6. What is the probability of getting heads exactly 3 times if you flip a fair coin 6 times?
b) A bag of marbles contains 10 red ones and 10 blue ones. Not a Binomial random variable c) You draw 10 of them; X = the number of red ones. X is Binomial random variable.
Q6. x = number of successes = 3 n = number of trials = 6 p = probability of success = .5 q = probability of failure = 1-p = 1 - .5 = .5 Probability of getting 3 heads = P(X = 3) = C(6, 3) (.5)3(.5)6-3 = 20*.125 * .125 = .3125
Q7. Your name is Edison and you are an expert penalty goal shooter. your skill has been improved for the past 10 years, and now you are as good as you will ever be. Your success rate has been measured at 80%. Thus, p = .8 and q = .2. You take n = 6 shots on goal, so the possible values of X (the number of successes) are 0, 1, 2, 3, 4, 5, 6. Here is the probability for each value of X:
P(X = 0) = C(6, 0) (.8)0(.2)6 = 1 * 1 * .000064 = .000064 Putting them all together gives the probability distribution for
x 0 1 2 3 4 .2457 6 5 6
X:
.39321 .26214 6 4
P(X = 0) = C(6, 0) (.8)0(.2)6 = 1*1 *.000064 = .000064 P(X = 1) = C(6, 1)(.8)1(.2)5 = 6 * (.8)1*(.2)5 = .001536 P(X = 2) = C(6, 2) (.8)2(.2)4 = 15 * (.8)2 * (.2)4 = .001536 P(X = 3) = C(6, 3) (.8)3(.2)3 = 20 * (.8)3 * (.2)3 = .08192 P(X = 4) = C(6, 4) (.8)4(.2)2 = 15* (.8)4* (.2)2 = .24576 P(X = 5) = C(6, 5) (.8)5(.2)1 = 6 * (.8)5 * (.2)1 = .393216 P(X = 6) = C(6, 6) (.8)6(.2)0 = 6 * (.8)6 * (.2)0 = . .262144