Probability Distributions. Statistical Experiments – any process by which measurements are...
-
Upload
linda-hancock -
Category
Documents
-
view
217 -
download
1
Transcript of Probability Distributions. Statistical Experiments – any process by which measurements are...
Probability Distributions
Statistical Experiments – any process by which measurements are obtained.
A quantitative variable x, is a random variable if its value is determined by the outcome of a random experiment.
Random variables can be discrete or continuous.
Statistical Experiments and Random Variables
Discrete vs. Continuous Discrete random variables – can take
on only a countable or finite number of values. (Countable)
Continuous random variables – can take on countless values in an interval on the real line. (Measurable)
Which measurement involves a discrete random variable?
a). Determine the mass of a randomly-selected penny
b). Assess customer satisfaction rated from 1 (completely satisfied) to 5 (completely dissatisfied).
c). Find the rate of occurrence of a genetic disorder in a given sample of persons.
d). Measure the percentage of light bulbs with lifetimes less than 400 hours.
Discrete or Continuous?Gas mileage(mpg) of all 2014
Toyota Prius’Number of songs on a seniors
iPod1 mile time of a high school
juniorAmount of m&ms in a bagHeight of a high school freshman
Probability distributions An assignment of probabilities to the
specific values or a range of values for a random variable.
Discrete Probability Distribution1) Each value of the random variable
has an assigned probability.
2) The sum of all the assigned probabilities must equal 1.
Discrete Probability Distribution Rolling a die
P(1) = .167, P(2) = .167, P(3) = .167…
# on the Die0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
123456
Discrete Probability Distribution
15 16 17 18 190
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Age of AP Statistics Students
Age of AP Statistics Students
Discrete Probability Distribution
# of cars
0 1 2 3
P(x) .08 .72 .18 .02
We asked 200 adults how many cars they own. The results are shown as a probability distribution below.
1) What is the probability of someone owning 2 cars?
2) What is the probability of someone owning at least 1 car?
3) What is the probability of someone owning no more than 1 car?
What would a histogram look like if we graphed the weights (to the nearest pound) of 100 newborn babies?
What would it look like if we graphed the same 100 newborn babies, however rounded the weights to the nearest tenth?
What about hundredth?
Probability Distributions for Continuous Random Variables
Probability Distributions for Continuous Random VariablesThe probability distribution is
given as a density curve.The function that defines this
curve is called the density function f(x)
Probability Distributions for Continuous Random VariablesWhen the density is constant
over an interval, the probability distribution is called a uniform distribution.
Probability Distribution Features
Since a probability distribution can be thought of as a relative-frequency distribution for a very large n, we can find the mean and the standard deviation.
When viewing the distribution in terms of the population, use µ for the mean and σ for the standard deviation.
Mean and Standard Deviationof Discrete Probability Distribution
(Expected Value)
Find the mean and standard deviation of the following…
x 0 1 2 3 4 5 6 7 8 9 10
p(x)
.002
.001
.002
.005
.02 .04 .17 .38 .25 .12 .01
µ = 7.16,
1) What is the probability that a child receives a score within 2 standard deviations of the mean?
Sometimes we are interested in the behavior of not just a random variable itself, but a function of the variable.Suppose we know the mean and
standard deviation of the number of gallons of propane a random person orders in Old Lyme to be 318 and 42 respectively.
A company is considering 2 different pricing models:
Model 1: $3 per gallonModel 2: service charge of $50 +$2.80 per gallon
Mean and Standard Deviation of Linear FunctionsHelps us understand the behavior of
functions of random variables. Mean of y = a + bx is
a + bStandard deviation of y is
Which pricing model is better?x = # of gallons of propane
Model 1= $3.00xModel 2 = $50 + $2.80 per gallonWe are interested in y(amount
billed)Find the mean and standard
deviation of both models and compare them.
Mean and Standard Deviation of Linear Combinations
(must be independent)
If the free response is given twice as much weight, what are the mean and s.d of y?
Mean Standard Deviation
Mult. Choice 38 6
Free Response 30 7
Binomial Experiment (2 outcomes)1) There are a fixed number of trials.
This is denoted by n.2) The n trials are independent and
repeated under identical conditions.3) Each trial has two outcomes:
S = success F = failure4) For each trial, the probability of
success, p, remains the same. Thus, the probability of failure is 1 – p = q.
Binomial ProbabilityThe central problem is to determine
the probability of x successes out of n trials.◦Mark is an 80% free throw shooter. What
is the probability that he makes exactly 5 out of 10 free throws?
◦What is the probability he makes exactly 10 out of 10?
◦What is the probability he makes more than half his free throws?
Probability of x successes
Find the probability of observing 6 successes in 10 trials if the probability of success is p = 0.4.
a). 0.111 b). 0.251 c). 0.0002 d). 0.022
Using the Binomial Table
1) Locate the number of trials, n.
2) Locate the number of successes, x.
3) Follow that row to the right to the corresponding p column.
Find the probability of observing 3 successes in 5 trials if p = 0.7
Finding the Probability of multiple successes, r’s.
Find the probability of observing less than three 3 successes in 5 trials if p = 0.7
P(r<3) = P(0) + P(1) + P(2)
Find the probability of observing less than 3 successes in 5 trials if p = 0.7
Graphing a Binomial Distribution
Same as a relative frequency histogram with r values on the horizontal axis.
Graph n=3, and p=.2
# of Successes0
0.1
0.2
0.3
0.4
0.5
0.6
r=0r=1r=2r=3
Mean and Standard Deviation
# of Successes0
0.1
0.2
0.3
0.4
0.5
0.6
r=0r=1r=2r=3
Graph n=3, and p=.2
𝜇=.6𝜎= .6928
Think about it!The graph of a distribution with a
small probability of success(p) would be skew left, right, symmetrical, uniform or bimodal?
How about a probability of success being 0.5?
The Normal Curve (Bell Curve)
Finding z scores
z score gives the number of standard deviations between the x value and the mean µ.
z =
Always round to the hundredths.
Finding area (probability) using z score and the standard NORMAL DISTRIBUTION
Table 2 gives you the area to the left of z.
This is the probability of less than z.
Area to the right of z
Area Between two z values
Inverse Normal distributionWe have been working with
finding an area (probability), given a certain x or z value.
We can also find an x or z value given a certain area (probability).
Left-tail case: the given area is to the left of z
Look up the number A in the body of the table and use the corresponding z value.
Right-tail case: the given area is to the right of z
Look up the number 1 – A in the body of the table and use the corresponding z value.
Center-tail case: the given area is symmetric and centered above z = 0.
Look up the number (1 – A)/2 in the body of the table and use the corresponding ±z value.
Using Table 3 in the Appendix, find the range of z scores, centered about the mean, that contain 70% of the probability.
a). –1.04 to 1.04 b). –2.17 to 2.17
c). –0.30 to 0.30 d). –0.52 to 0.52
Checking for NormalityNormal Probability Plot: the
scatterplot of (normal score, observed value)
If the normal probability plot shows a linear trend, then the data is approximately normal.
Using the Correlation Coefficient to Check NormalityCorrelation Coefficient is obtained
using (normal score, observed value)If r is close to 1, then a linear
relationship should be represented. How close to 1 is close to 1?If r is less than the critical r for the
corresponding n, it is not reasonable to assume a normal distribution
If n is between two sample sizes, use the larger n
Approximating Discrete/Binomial Distributions using Normal Dist.
Approximating Discrete/Binomial Distributions using Normal Dist.
How do we tell that a binomial distribution is normal?
If np ≥ 10 and nq ≤ 10, then the distribution is approximately normal.
Add or subtract .5 to your x-values and compute the probability as you would a normal distribution.
A Biologist found that the probability is only 0.65 that a given Arctic Tern will survive the migration from its summer nesting are to its winter feeding groups. A random sample of 500 Artic Terns were branded at their summer nesting area. Find the probability that between 310 and 340 of the branded Artic Terns will survive the migration.