L4&5- hypothesis testing Flashcards

(13 cards)

1
Q

reject H0 when

A

z > z* ; |test stat| > |cv|

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

test stat for large sample

A

z = (p-𝜋) / [𝜋(1-𝜋)/n]

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

test stat general formula

A

(sample est) - (hypothesises val) / SE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

if test stat < cv

A

not enough evidence to reject H0 (decision rule)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

if sample size is small

A

t distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

test stat for t distr

A

t = (x̄ - μ) / [√s^2 / n)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

if |test stat| < |t*|

A

cannot reject H0 (decision rule)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

method

A
  • H0 and H1
  • sign level -> alpha / 2
  • look up cv (z* or t*)
  • compute test stat
  • decision rule
  • sketch
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

chi squared distr characteristics

A

asymmetric, non negative (no +/-), depends on v

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

when is z / t used

A

testing for means

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

when is chi squared used

A

estimating pop var (point & interval estimate)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

f table F*

A

⍺/2, v, v = F*

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

calculating test stat

A

larger variance on top

How well did you know this?
1
Not at all
2
3
4
5
Perfectly