Interval Estimation Flashcards

1
Q

Define a confidence interval

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

Give the equation for S2

A

Note: We assume the sample comes from a normal distribution. S2 is a random variable as it is a transformation of the random sample variables

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

Give the equation for x̄ (X bar)

A

Note: we assume the sample comes from a normal distribution. X bar is a random variable as it is a transformation of the random sample variables

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

Give the test statistic needed to construct a (1-α) confidence interval for unknown mean µ and we know population variance σ2

A

This can be rearranged to find µ

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

Give the test statistic needed to construct a (1-α) confidence interval for unknown mean µ and we do not know population variance

A

Note: we can rearrange T to find µ

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

Describe the width of an approximate confidence interval for the population mean

A

Note: if σ is unknown it can be replaced by S (the sample standard deviation)

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

Describe how to find the confidence interval for a population mean if the distribution is not necessarily normal and we know the variance.

A
  • Using central limit theorem, all IIDs of any distribution are approximately normal. Generally, n≥30 is sufficient to make this assumption.
  • Variance is often a function of E(X) (Ie one parameter distributions) and so T can be modified for different distributions. See below
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Describe how to find the confidence interval for a population mean if the distribution is not necessarily normal and we do not know the variance.

A

Note that this approximation is valid whenever CLT is valid.

As we make no assumptions we must be sure that the sample is very large for the CI to be valid.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Give the test statistic needed to construct a (1-α) confidence interval for unknown variance σ2 and we know the population mean µ.

A

This can be rearranged for σ2

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

Give the test statistic needed to construct a (1-α) confidence interval for unknown variance σ2 and we don’t know the population mean µ.

A

This can be rearranged for σ2

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

Give the test statistic needed to construct a (1-α) confidence interval for the difference in means between two independent normal distributions and where we know the variance

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

Give the test statistic needed to construct a (1-α) confidence interval for the difference in means between two independent normal distributions and where we don’t know the variance

A

We assume the variances of the two distributions are equal.

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

Give the test statistic needed to construct a (1-α) confidence interval for the difference in means where we have one sample of pairs

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Describe the random sample of differences

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Give the test statistic needed to work out the confidence interval for the ratio of variances from two normal populations when the samples are independent and we know the means

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Give the test statistic needed to work out the confidence interval for the ratio of variances from two normal populations when the samples are independent and we don’t know the means

A
17
Q

Define a simple hypothesis

A

A hypothesis is called simple if it completely determines the distribution of the observed data; an example of a simple hypothesis is θ = θ0 if θ is the only parameter in the model.

18
Q

Define a composite hypothesis

A

The hypothesis is called composite if it contains an unknown parameter. an example is a hypothesis of the form θ > θ0, since it does not exactly determine all the parameters.

19
Q

Define a type 1 error

A

H0 is rejected even though it is true

20
Q

Define a type 2 error

A

H0 is not rejected even though it is false

21
Q

Define the level of significance for a hypothesis test

A
22
Q

Define the power of a test

A
23
Q

Give the equation for Sp2

A