17 Flashcards
P(- #sigma < x - mu < #sigma) is equal to?
P(- # < (x - mu)/sigma < #) where (x - mu)/sigma is standard normal.
(x - mu)/sigma_x is?
Z meaning standard normal
!!!As the sample size increase, the sample mean gets close to?
The actual mean of the population.
Confidence level?
How confident you are that an estimator falls within a internal estimate.
Interval estimate?
Range of numbers determined by following some kind of random procedure that is used to estimate a population parameter.
Confidence interval?
An interval estimate produced by following some procedure that will correctly estimate a population parameter at least 100(1- alpha)% of time. P(Confidence Interval surrounds parameter) = 1 - alpha
Margin of error (E)?
Maximum absolute difference between a parameter (theta) and 100(1 - alpha)% of all possible estimates (theta^_n)
Margin of error of X bar and mu if alpha = 0.95?
The two values are at most 2 standard deviations away 95% of the time.
Steps of making a confidence interval?
•determine minimum sample size n needed to create an interval with a desired confidence level 100(1 - alpha)% and desired margin of error (E_desired)
•collect your data by taking sample size n from population.
•using the sample data, we will compute the point estimate (theta^_n) of unknown parameter (theta)
•Use sample to compute actual margin of error E
•taking point estimate and adding and subtracting it. This will find the upper and lower bound.
•Certain percent confidence that lower bound is less than known parameter and that upper bound is larger than unknown parameter and some percent confidence that interval (lower, upper bound) captured the unknown parameter.
1-mean z interval needs what?
Assumptions to be true so that C.L. Is correct.
Assumptions of 1-mean z-interval? 2•
•population distribution is normally distributed with a known variance sigma^2
•a random sample of size n has been collected.