*Binomial Distributions Flashcards
What is a Bernoulli Distribution?
Any probability distribution with only 2 possible results, the desire result (x=1) and the undesired result (x=0)
In what 3 ways Bernoulli Distribution be represented?
- Piecewise function, the thing with the { and the two functions
- Probability Distribution Table, a table with two rows and three columns top row is x-values, bottom row is probability values
- Probability Bar Graph, x-values on the bottom, probability values on the side
What is p?
The probability of the desired result
What is q?
The probability of the undesired result
OR
q = 1 - p
Given that nothing extra is said, what is the mean?
E(x) = p
Given that nothing extra is said, what is the Variance?
V(x) = pq
OR
V(x) = p(1 - p)
What is the standard deviation and how is it found?
√V(x) = √(pq) = σ
The spread of the data
Given that we are told the number of trials (n) and the number of wanted results (r), the probability can be found by-
Binomial Formula: nCr • p^r • q^(n - r)
Drawing a probability histogram
1. First step?
Locate the n and the p
Make r and q
E.g. Draw a histogram for the probabilities of rolling a 2 on a dice 5 times
n = 5, p = 1/6
If this is the case,
• r = 0, 1, 2… n
• q = 5/6
Drawing a probability histogram
2. Second step?
Find probabilities with the Binomial Formula for all values (0 → n = r-values)
E.g. Draw a histogram for the probabilities of rolling a 2 on a dice 5 times
Do nCr p^r q^(n-r) for r = 0, 1, 2, 3, 4, 5
Drawing a probability histogram
3. Third step?
Put r-values on the x-axis, put probability on the y-axis
E.g. Draw a histogram for the probabilities of rolling a 2 on a dice 5 times
Draw the graph
Drawing a probability histogram
4. Fourth step?
Label its shape, check if it’s a slide
If there’s isn’t a slide and it’s even → symmetrical
If there is a slide
- If it’s fun (short to long) → positive skew
- If it’s not fun (long to short) → negative skew
E.g. Draw a histogram for the probabilities of rolling a 2 on a dice 5 times
Here, there’s no slide and it’s even, thus symmetrical
Given that we’re given a number of trials, what is the mean?
E(x) = np
Given that we’re given a number of trials, what is the variance?
V(x) = npq
OR
V(x) = np(1 - p)
note with the same standard deviation formula √V(x)
A sample population is?
A small portion of the total population, given as s
An observation (desired result) is?
The number of those with a desired trait in the small population, given as d
A sample proportion is?
The probability of finding a desired trait within a small portion, given as p^ = d/s
Given that we’re told it’s a sample, what is the mean?
E(x) = p = p^ = d/s
Given that we’re told it’s a sample, what is the variance?
V(x) = p^q^/s
Finding the probability of a sample proportion
1. First step?
Find the z score, approximate to 2 decimals
Finding the probability of a sample proportion
2. Second step?
Check the table for the t-score:
If it’s less than question → straight to the t-score
If it’s greater than question → 1 minus the t-score
If it’s between question → bigger t-score minus the smaller t-score
- note* table is in page 633-34
- note* only works for normal distributions
A z-score can be found with the formula, and WHATS so special about this one (binomials)?
z = (x - μ)/σ
x = probability of a given event happening μ = p^ = d/s σ = √(p^q^/s)
If we are given p^, what is q^?
q^ = 1 - p^