HP 50G

Page 18-22
L @)STAT @PLOT @SCATR produce scattergram of y vs. x
@STATL show line for log fitting
Θ To return to STAT menu use: L@)STAT
Θ To get your variable menu back use: J.
Confidence intervals
Statistical inference is the process of making conclusions about a population
based on information from sample data. In order for the sample data to be
meaningful, the sample must be random, i.e., the selection of a particular
sample must have the same probability as that of any other possible sample out
of a given population. The following are some terms relevant to the concept of
random sampling:
Θ Population: collection of all conceivable observations of a process or
attribute of a component.
Θ Sample: sub-set of a population.
Θ Random sample: a sample representative of the population.
Θ Random variable: real-valued function defined on a sample space. Could
be discrete or continuous.
If the population follows a certain probability distribution that depends on a
parameter θ, a random sample of observations (X
1
,X
2
,X
3
,... , X
n
), of size n,
can be used to estimate θ.
Θ Sampling distribution: the joint probability distribution of X
1
,X
2
,X
3
,... , X
n
.
Θ A statistic: any function of the observations that is quantifiable and does not
contain any unknown parameters. A statistic is a random variable that
provides a means of estimation.