Textbook Notes (280,000)
US (110,000)
UCLA (2,000)
STATS (30)
STATS 13 (10)
Tsiang, Mike (2)
Chapter 16, 10
STATS 13 Chapter Notes  Chapter 16, 10: Statistical Inference, Statistic, Sampling Distribution
by OC2737311
Department
StatisticsCourse Code
STATS 13Professor
Tsiang, MikeChapter
16, 10This preview shows pages 13. to view the full 27 pages of the document.
ph
introduction
â€¢
Steps
of
a
stats
investigation
:
i
.
ask
a
research
Question
that
can
be
answered
by
data
2.
design
a
study
and
collect
data
the
HOW
3
.
explore
data
patterns
4
.
draw
inferences
from
data
is
the
pattern
reliable
?
(
sample
size
)
5
.
formulate
conclusions
a
prude
o
no
Negara
una
conclusion
?
g.
look
back
and
ahead
limitations
,
im
pgÂµme
â€¢
Statistical
inference
pillars
:
effect
1.
Significance
:
how
singer
is
the
evidence
?
pattern
2.
estimation
:
what
is
the
size
of
the
effect
?
3.
generalization
:
how
broadly
do
conclusions
apply
?
4.
Causation
:
can
we
determine
the
Caza
?
â€¢
Basic
terminology
:
â€¢
data
:
values
measured
/
categories
â€¢
observational
units
:
individual
entities
on
which
data
are
recorded
eg
.
participants
*
variables
:
measured
quantities
>
quantitative
:
numbers
>
categorical
(
qualitative
)
:
categories
â€¢
distribution
:
pattern
of
outcome
Only pages 13 are available for preview. Some parts have been intentionally blurred.
Pt
exploring
data
â€¢
variability
:
how
predictable
is
a
certain
process
leituatnst
â€¢
shape
of
dist
.
:
bimodal
?
kÂ¥1
?
"
d
"
'
tht
dat
"
"
"
"
ha
"
7Â¥
mode
?
Â¥he
center
of
the
distribution
is
the
mean
,
nuhich
is
not
always
the
mode
>
Variability
is
measured
by
the
standard
deviation
t
variance
how
far
is
the
data
from
the
mean
â€¢
Unusual
observations
:
outliers
don't
fit
pattern
pre
random
processes
difficult
to
predict
exactly
opposite


deterministic
â€¢
random
processes
the
chances
of
any
of
the
events
/
possibilities
outcome
always
happening
is
the
same

he
can
be
repeated
the
same
infinitive
by
.
â€¢
probability
:
proportion
of
outcomes
=
expected
in
the
long
run
total
â€¢
probability
model
:
assumption
of
how
random
processes
may
generate
data
.
Only pages 13 are available for preview. Some parts have been intentionally blurred.
1
significance
â€¢
is
the
effect
evidently
caused
by
the
suspected
cause
,
or
are
there
other
factors
?
(
on
,
strong
is
the
relationship
of
the
variables
4
the
data
?
>
how
do
we
know
if
results
were
caused
by
chance
?
â†³
null
hypothesis
:
there
is
no
relation
us
up
.
hypothesis
:
there
is
a
relation
I
are
results
statistically
significant
?
7
.
7
chance
models
â€¢
are
the
Agata
points
far
apart
enough
to
be
statistically
significant
?
x
what
is
the
probability
of
.
.
.
if
random
event
?
Wil
compare
the
results
to
the
hull
â†’
Karen
doffs
)
t
what
is
the
actual
probability
?
within
a
J
.
Sample
:
set
of
observational
units
that
the
data
is
collected
from
.
relevant
population
â€¢
statistic
:
number
I
percentage
that
summarizes
results
.
eg
.
"
416
upon
P
of
PM
.
sample
size
:
#
of
sets
of
data
.
eg
.
#
trials
â€¢
parameter
:
long
run
numerical
property
of
a
process
.
â€¢
STATISTIC
't
PARAMETER
otsttammpu
emoting
went
priosbaabi
lily
parameter
You're Reading a Preview
Unlock to view full version