Chapter 5 (Continuous Probability Distributions) Flashcards
What are some facts about the Normal Distribution?
- it’s continuous between -infinity and + infinity
- there are two parameters: myu and the variance (sigma^2)
- curve is bell shaped
- symmetric around myu
- it’s very common
- pdf is not an integral by hand - use a table to find areas under the curve.
Why is the normal distribution so common?
1) lots of populations have normal distributions.
2) sims of means of random variables have normal distributions under certain conditions regardless of the underlying population distribution.
How do we denote the Normal Distribution?
denoted: X~N(mu, sigma^2)
What is the normal curve rules of thumb?
68 - 95 - 99.7 Rule!
68% lie within 1 SD
95% lie within 2 SDs
99.7% lie within 3 SD/
How do you find a “z-score”?
X-muu
–––––––
sigma
Or
Value - mean / standard deviation
How to notate P(Z ≤ z) with phi?
Phi(z)
Sums of independent, normally distributed random variables are…
Also normally distributed.
If many random variables Xi are independent, then Xbar has what distribution?
Xbar ~ N(mu, sigma^2/n)
What is the central limit theorem?
If X_1, X_2, …, X_n, is a sample (ind. random variables from one population) from some distribution with mean muy and variance sigma^2, if the sample size n is large enough, the distribution of their average is approximately normal:
X_bar~N(myu, sigma^2/n)
What does “iid” mean?
Independent and identically distributed (cane from same population (distribution)).
What is the distribution of mean (X_bar) according to the central limit theorem?
X_bar~•N(mu, sigma^2/n)
~• is approximately distributed
What is the sum (nX_bar) according to the central limit theorem?
nX_bar ~• N(n * mu, n*sigma^2)
What is the proportion (p_hat) according to the central limit theorem?
P_hat ~• N(p, p*(1-p)/n)
What are the conditions for the CLT?
1 and 2) the random variables are iid. (Independent and identically (same)distributed)
3) the sample size n is large enough (usually 25+)
4) Were concerned with the mean (X_bar), sum (n*X_bar), or proportion (p_hat)
Under certain circumstances a Binomial distribution may look like a normal distribution. What are these circumstances?
X•~ N(np, np(1-p))
Where np ≥ 5 and np(1-p) ≥ 5