Model Misspecification

Question 1

Model Misspecification - Omitted variable

Answer 1

If we omit a significant variable from our model, the error term will capture the missing.

Question 2

Model Misspecification - Inappropriate form of variable

Answer 2

Failing to account for non-linearity
Causes: Conditional heteroscedasticity

To fix it we can use natural log to transform the variable to be linear.

Question 3

Model Misspecification - Inappropriate Scaling

Answer 3

Causes Conditional heteroscedasticity and multicollinearity

Question 4

Model Misspecification - Inappropriate Pooling of Data

Answer 4

Causes Conditional heteroscedasticity and Serial correlation

• Conditional heteroscedasticity
• Serial Correlation

Question 5

What is Unconditional heteroscedasticity

Answer 5

Var(error) not correlated with independent X variable.
No issue with interference.

Question 6

What is Conditional heteroscedasticity

Answer 6

Var(error) are correlated with independent X variable

F-test is unreliable since MSE is a biased estimator of the true population variance.

variance at one time step has a positive relationship with variance at one or more previous time steps. This implies that periods of high variability will tend to follow periods of high variability and periods of low variability will tend to follow periods of low variability.

Question 7

What does the Breusch Pagan BP tets do?

Answer 7

Tests for heteroskedasticity

Question 8

The formula for BP test statistics

Answer 8

n * R-Square

Question 9

BP test
Test statistics > Critical value

Answer 9

Reject the null.
No heteroskedasticity
homoskedasticity is present - Constant vartiance

• H0: No heteroskedasticity - homoskedasticity is present
• Ha: Heteroskedasticit

Question 10

BP test
Test statistics < Critical value

Answer 10

Reject the null

There is Heteroskedasticity

• H0: No heteroskedasticity
• Ha: Heteroskedasticity

Question 11

What is serial correlation?

Answer 11

Errors correlated across the observation

Question 12

Positive Serial Correlation

Answer 12

Positive residuals is most likely followed by positive residuals
Negative residuals is most likely followed by negative residuals

Question 13

Negative Serial Correlation

Answer 13

Negative residual is most likely followed by positive residual
Positive residual is most likely followed by negative residual

Question 14

Multicollinearity

Answer 14

2 or more independent variables are highly correlated or there is an approximate linear relationship among the IVs.

Inflates the standard error

Coefficients will be consistent but imprecise and unreliable
Inflated SE and insignificant T-Statistics, but possibly significant F-Statistics

Question 15

How to detect multicollinearity?

Answer 15

Variance inflation factor

1 / (1- R Square)

We want VIF as low as possible

> 5 Concerning
10 Multicollinearity

Question 16

How to fix multicollinearity?

Answer 16

• Increase sample size
• Excluding one or more of the regression variables.
• Use a different proxy for one of the variables

Question 17

What does the BP test for ?

Answer 17

Conditional Heteroskedasticity

Question 18

What is the most common problem in trend models?

Answer 18

Serial correlation

Trend models often have the limitation that their errors are serially correlated. This is due to the fact that predictions in the trend models are based soley on what time period it is, and thus they fail to account for significant trends in the data such as recession.

Question 19

Chi-Squar degree of freedom > BP

Answer 19

No Evidence of Conditional Heterskedatisicty

Question 20

BP test > Chi-Squar Degree of Freedom

Answer 20

Evidence of Conditional Heterskedatisicty

Question 21

Spurious Correlation

Answer 21

Two or more events or variables are associated but not causally related, due to either coincidence or the presence of a certain third, unseen factor.

Caused by innapropriately pooled data.

Question 22

Characteristics of Multicolinearity

Answer 22

Even though we have a independent variable that is not statistically significant.

High R-Square
Significant F - Statistics