ADV STAT MIDTERM ASSUMPTIONS

Question 1 Assumptions:

1. Linearity

2. Normality of residuals

3. Homoscedasticity

4. Independence of errors

5. Multicollinearity

6. Outliers and influential observations

7. Appropriate variables and sample size

Question 1 Assumptions:

1. ———-

2. Normality of residuals

3. Homoscedasticity

4. Independence of errors

5. Multicollinearity

6. Outliers and influential observations

7. Appropriate variables and sample size

Linearity

Question 1 Assumptions:

1. Linearity

2. ——————

3. Homoscedasticity

4. Independence of errors

5. Multicollinearity

6. Outliers and influential observations

7. Appropriate variables and sample size

Normality of residuals

Question 1 Assumptions:

1. Linearity

2. Normality of residuals

3. ————-

4. Independence of errors

5. Multicollinearity

6. Outliers and influential observations

7. Appropriate variables and sample size

Homoscedasticity

Question 1 Assumptions:

1. Linearity

2. Normality of residuals

3. Homoscedasticity

4. ———————

5. Multicollinearity

6. Outliers and influential observations

7. Appropriate variables and sample size

Independence of errors

Question 1 Assumptions:

1. Linearity

2. Normality of residuals

3. Homoscedasticity

4. Independence of errors

5. ———————-

6. Outliers and influential observations

7. Appropriate variables and sample size

Multicollinearity

Question 1 Assumptions:

1. Linearity

2. Normality of residuals

3. Homoscedasticity

4. Independence of errors

5. Multicollinearity

6. ———————-

7. Appropriate variables and sample size

Outliers and influential observations

Question 1 Assumptions:

1. Linearity

2. Normality of residuals

3. Homoscedasticity

4. Independence of errors

5. Multicollinearity

6. Outliers and influential observations

7. ————————

Appropriate variables and sample size

Q1 Assumption:

1. Linearity:

Is the relationship approximately straight?

Q1 Assumption:

2. Normality of residuals:

Do the errors look bell-shaped?

Q1 Assumption:

3. Homoscedasticity:

Is the error spread consistent?

Q1 Assumption:

4. Independence of errors:

Are the errors unrelated?

Q1 Assumption:

5. Multicollinearity:

Are predictors too similar?

Q1 Assumption:

6. Outliers and influential observations:

Is one observation driving the results?

Q1 Assumption:

7. Appropriate variables and sample size:

Do I have appropriate data and enough of it?

Q1 Assumptions - LINEARITY:

- What I Checked in SPSS:

Residual plot — no obvious curved pattern.

Q1 Assumptions - Normality of residuals:

- What I Checked in SPSS:

Histogram and P–P plot — approximately normal.

Q1 Assumptions - Homoscedasticity:

- What I Checked in SPSS:

Residual scatterplot — no obvious funnel shape.

Q1 Assumptions - Independence of errors:

- What I Checked in SPSS:

Durbin–Watson = 1.968 (close to 2 is good). Independence also depends on study design.

Q1 Assumptions - Multicollinearity:

- What I Checked in SPSS:

VIF = 1.088–1.222 (below 5 is good).

Q1 Assumptions - Outliers / Influence:

- What I Checked in SPSS:

Standardized residuals = −2.866 to 2.829 (within ±3). Max Cook's D = .132 (below 1; may warrant closer inspection).

Q1 Assumptions - Appropriate variables / Sample size:

- What I Checked in SPSS:

Quantitative variables; N = 200, four predictors.