lecture 3. statistical inference Flashcards
does this describe exploratory data analysis or statistical inference?
its purpose is unrestricted exploration of data, searching for interesting patterns; conclusions apply to the subjects and circumstances for which we have data in hand; conclusions are informal based on what we see in the data
exploratory data analysis
does this describe exploratory data analysis or statistical inference?
purpose is to answer a specific question, posed before the data produced; conclusions apply to a larger group of subjects or a broader class of circumstances; conclusions are formal, backed by a statement of our confidence in them
statistical inference
this is a method for drawing conclusions about a population from a sample. uses probability to indicate how trustworthy its conclusions are. assumes a random sample - if this is not the case your conclusions may be faulty
statistical inference
what are the two most common types of statistical inference?
- significance tests
- confidence intervals
this common type of statistical inference estimates a population parameter.
confidence intervals
this common type of statistical inference assess the evidence in the data for some claim about the population
significance tests
generating statistics from a relatively small sample to provide an indication of a population value (parameter) is the process of ________
estimation
________ are fixed for a given population (the mean for a given population is constant)
parameters
_______ are estimates of parameters and cary sample by sample. in effect they are random variables.
sample statistics
the ____________ distribution is the distribution of all possible values of that statistic in all possible random sample so the same size n from population N
random sampling
if a random variable x has a population mean μ and population variance σ2, then the sampling distribution of means (of samples of size n) will have a mean of μ and variance ____
σ2/n
this property means variability of the random sampling distribution depends on _______ and ________
sample size and the variability of the population
the ______ (smaller/larger) the sample size, the smaller the σ2/n
larger
_______ (more/less) variance = the more certain we can be
less
since sample means are normally distributed, we can define a ________ which is a range of values used to estimate the true value of the population parameter; it is the probability (usually expressed as a percentage) or the proportion of times that the ____ actually does contain the population parameter
confidence interval (CI)