Question

Help with some data science questions

Q.1 The linear regression model assumes multivariate normality, no or little multicollinearity, no auto-correlation, and homoscedasticity? Which assumption is missing from this list? (no more than 10 words)

Q.2 The coefficient of correlation measures the percent change in the feature variables explained by the target variables.

a) True

b) False

Q.3 In a linear regression model, the coefficient measures the change in Y explained by one unit-change in X.

a) True

b) False

Q4. The normal distribution has a kurtosis coefficient of 3.

a) True

b) False

Q5. Under-fitting means that a linear regression model is not able to capture the patterns of data.

a) Ture

b) False

Q6. If a visual inspection of the data shows a curvilinear relationship, is it appropriate to use a polynomial regression model?

a) Yes

b) No

Q7. A polynomial regression model is robust to outliers.

a) True

b) False

Q8. When working with different models, the one with the highest AIC or BIC is preferred.

a) True

b) False

Q9. You cannot use the least-square method to create a line that would best fit the data.

a) True

b) False

10. The cost function is a measure of how wrong the model is in terms of its ability to estimate the relationship between X and Y.

a) True

b) False

Answer 1

2. The coefficient of correlation measures the percent change in the feature variables explained by the target variables. (False)

3.  In a linear regression model, the coefficient measures the change in Y explained by one unit-change in X. (True)

4. The normal distribution has a kurtosis coefficient of 3 (True)

5.  Under-fitting means that a linear regression model is not able to capture the patterns of data. (True)

6. If a visual inspection of the data shows a curvilinear relationship, it is appropriate to use a polynomial regression model. (True)

7. A polynomial regression model is robust to outliers. (True)

8. When working with different models, the one with the highest AIC or BIC is preferred. (False)

9. You cannot use the least-square method to create a line that would best fit the data. (False)

10. The cost function is a measure of how wrong the model is in terms of its ability to estimate the relationship between X and Y. (True)