Topic 5: Misspecifications

Question 0

Omitted Relevant Variable

- what are the consequences?

Answer 0

Biased coefficient estimates of the included variables correlated with the omitted ones

Question 1

Omitted Relevant Variable

- what is it and what causes it?

Answer 1

Variable that is correlated with the included variables but not included in the model

Question 2

Omitted Relevant Variable

- how to detect it?

Answer 2

• theory
• significant unexpected signs
• RESET test

Question 3

Omitted Relevant Variable

- remedy?

Answer 3

Include omitted variable or a proxy

Question 4

Irrelevant Variable

- what is it and what causes it?

Answer 4

The inclusion of an unnecessary variable

Question 5

Irrelevant Variable

- what are the consequences?

Answer 5

Lowers precision of model
• inflated standard errors
• low t-ratios

Question 6

Irrelevant Variable

- how to detect it?

Answer 6

• theory
• t-test on beta
• adjusted r^2 increases if variable is dropped

Question 7

Irrelevant Variable

- remedy?

Answer 7

Exclude the irrelevant variable

Question 8

Incorrect Functional Form

- what is it and what causes it?

Answer 8

The functional form of the model might not be linear

Question 9

Incorrect Functional Form

- what are the consequences?

Answer 9

• biased and inconsistent estimates

* poor fit of model (low R^2)

Question 10

Incorrect Functional Form

- how to detect it?

Answer 10

• theory
• Ramsey RESET
• scatter plot of Y with each of the X’s

Question 11

Incorrect Functional Form

- remedy?

Answer 11

• transform data into logs to linearise model

* add higher order functions of the variables to capture curvature

Question 12

Multicollinearity

- what is it and what causes it?

Answer 12

When some of the explanatory variables are highly correlated with one another

Question 13

Multicollinearity

- what are the consequences?

Answer 13

• high R^2, coefficients high SEs -> low t-ratios
• regression sensitive to small changes
• wide confidence intervals for parameters, incorrect inferences from model

Question 14

Multicollinearity

- how to detect it?

Answer 14

• Correlogram

* see R^2 of regression of X on all other X’s

Question 15

Multicollinearity

- remedy?

Answer 15

• ignore if model is okay
• drop collinear variable
• transform correlated variables into a ratio
• collect more data (longer sample period, higher frequency obs)

Question 16

Autocorrelation

- what is it and what causes it?

Answer 16

Observations of the residuals are correlated over time
Causes: • omitted variables/common shocks
• Business Cycle inertia
• Overlapping effect of shocks
• Model misspecification

Question 17

Autocorrelation

- what are the consequences?

Answer 17

• unbiased but inefficient
• incorrect inferences
• inflated R^2

Question 18

Autocorrelation

- how to detect it?

Answer 18

• Durbin Watson
• Breusch Godfrey
• Correlogram of residuals
• Ljung-box

Question 19

Autocorrelation

- remedy?

Answer 19

• GLS (if form is known)
• Dynamic Models
• HAC coefficients
• SE Newey-West

Question 20

Heteroscedasticity

- what is it and what causes it?

Answer 20

Variance of error term not constant for all observations
Causes: • scale/size effects
• measurement error
• subpopulation differences
• flow of info is time varying

Question 21

Heteroscedasticity

- what are the consequences?

Answer 21

• unbiased but inefficient estimates

* end up drawing wrong conclusions from hypotheses testing because of incorrect standard errors

Question 22

Heteroscedasticity

- how to detect it?

Answer 22

• visual inspection of residual plot graph
• white’s test
• engle’s LM test for ARCH

Question 23

Heteroscedasticity

- remedy?

Answer 23

• GLS (if form is known)
• transform variables using logs
• white’s SE estimates

Question 24

Seasonality | - what is it and what causes it?

Answer 24

``` Observations of the dependent variable are systematically higher/lower in certain periods Causes: • day of the week effect • January effect • Bank holiday effect • open/close market effect ```

Question 25

Seasonality | - what are the consequences?

Answer 25

Serially correlated error

Question 26

Seasonality | - how to detect it?

Answer 26

Dummy variable for the period where the pattern is observed

Question 27

Seasonality | - remedy?

Answer 27

Intercept or slope dummy to account for seasonality

Question 28

Normality | - what is it and what causes it?

Answer 28

When the residuals are not normally distributed Causes: • outliers • Heteroscedasticity • Seasonality

Question 29

Normality | - what are the consequences?

Answer 29

* test statistics do not follow normal distribution * estimators not efficient * SEs are biased leading to wrong inferences

Question 30

Normality | - how to detect it?

Answer 30

* Bera-Jarque * histogram of residuals * skewness and jurros is

Question 31

Normality | - remedy?

Answer 31

* dummy variable to knock out outliers | * GARCH model

Question 32

Structural break | - what is it and what causes it?

Answer 32

Parameters are not constant over sample period

Question 33

Structural break | - what are the consequences?

Answer 33

Biased coefficient estimates

Question 34

Structural break | - how to detect it?

Answer 34

• chow test for structural break

Question 35

Structural break | - remedy?

Answer 35

* split the period | * dummy variables to account for different behaviour over the periods