Lecture 4- Model Fit Flashcards

1
Q

b1 estimates what

A

Parameter for a predictor

  • Direction of relationship/effect
  • Difference in means
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

b0 estimates what

A

The value of the outcome when predictors =0 (intercept)

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

Sums of squares represent

A

Total error

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

Because sums of squares are totals they can only be compared when

A

They are based on the same number of scores

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

df=

Degrees of freedom

A

N- p

Number of scores - number of parameters

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

What is SST

Total sum of squares

A

Total variability (variability between scores and the mean)

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

What is SSR

Residual sum of squares

A
  • Total residual/error variability (variability between the model and the observed data)
  • How badly the model fits (in total)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is SSM

Model sum of squares

A
  • Total model variability (difference in variability between the model and the grand mean)
  • How much better the model is at predicting Y than the mean
  • How well the model fits (in total)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Each sum of squared has

A

Associated degrees of freedom

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

The df is what in relation to SS

A

The amount of independent information available to compute SS (Sum of squares)

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

For each parameter estimated we lose

A

1 piece of independent information

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

To get SST we estimate

A

1 parameter

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

dfT=

A

N- p

Number of pieces of info - parameter

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

To get SSR we estimate

A

2 parameters (b0 and b1)

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

dfM=

A

dfT- dfR

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

The null model and the estimated model are distinguished by

A

b1, the slope

17
Q

SST=

18
Q

MS=

A

SS/df

Mean squared= sum of squares/ degrees of freedom

19
Q

Sums of squares errors can’t compared based on

A

Different amounts of information

20
Q

The average or mean squared error can be computed by

A

Dividing a SS by the amount of information used to compute it

21
Q

The df quantifies

A

The amount of information used to compute a sum of squared errors

22
Q

The F statistic is the ratio of

A

MSM to MSR

23
Q

If the model results in better prediction than using the mean then

A

MSM should be greater than MSR

24
Q

What r2 represent

A
  • The proportion of variance accounted for by the model

- The Pearson correlation coefficient between observed and predicted scores squared

25
What does adjusted r2 represent
An estimate of r2 in the population (shrinkage)
26
How to enter predictors when there is more than one predictors in a model
- Hierarchical - Forced entry - Stepwise
27
What is hierarchical entrance to predictors
- Experimenter decides the order in which variables are entered into the model - Best for theory testing
28
What is forced entry to predictors
All predictors are entered simultaneously
29
What is stepwise entrance to predictors
- Predictors are selected using their semi-partial correlation with the outcome - Can produce spurious results - Use only for exploratory analysis
30
SST is made up of
SSM and SSR
31
SSM is worked out by
Comparing the model to the null hypothesis | Slope- flat line