5. Bias Flashcards

1
Q

What are two types of graphs used for detecting outliers?

A
  1. Scatterplots (less helpful with several predictors)
  2. Histograms
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

In an average sample, 95% of standardised residuals should lie between…

A

+-2 SD

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

In an average sample, 95% of standardised residuals should lie between…

A

+-2.5SD

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

Any case for which the absolute value of the standardised residual is 3 or more, is likely to be…

A

An outlier

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

What does Cook’s distance measure?

A

Measures the influence of a single case on the model as a whole

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

In Cook’s distance, values greater than __ may be cause for concern

A

1

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

DF beta statistics tell us…

A

The change in b when a case is removed

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

With DF beta statistics, be wary of standardised values with absolute values…

A

> 1

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

The population model should have what two errors?

A
  1. Homoscedastic errors
  2. Independent errors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

The relationship between predictor(s) and outcome is…

A

Linear

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

The combined effect of predictors is…

A

Additive

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

A model’s errors refer to what?

A

The differences between predicted values and observed values of the outcome variable in the population model
These values cannot be observed

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

A model’s residuals refer to what?

A

The differences between predicted values and observed values of the outcome variable in the sample model
These values can be observed and are representative of the population model errors

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

What does “Errors should be independent” mean?

A

The population error in prediction for one case should not be related to the error in prediction for another case

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

What does “errors should be homoscedastic” mean?

A

Variance of population errors (residuals) should be consistent at different values of the predictor variable

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

bs are unbiased but not optimal. Standard error is incorrect. Therefore…

A

t-tests, p-values and confidence intervals will also be incorrect

17
Q

p-values associated with the bs of the model assume that…

A

the test statistic associated with them follows a normal distribution

18
Q

What are the three concepts of “The bootstrap”
1. Standard errors are derived empirically using a ____ technique
2. Results in robust ____ and ____
3. Designed for ____ samples (when normality matters)

A
  1. resampling
  2. confidence intervals, p-values
  3. small