Topic 5: Misspecifications Flashcards Preview

Econometrics > Topic 5: Misspecifications > Flashcards

Flashcards in Topic 5: Misspecifications Deck (36)
Loading flashcards...
0
Q

Omitted Relevant Variable

- what are the consequences?

A

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

1
Q

Omitted Relevant Variable

- what is it and what causes it?

A

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

2
Q

Omitted Relevant Variable

- how to detect it?

A
  • theory
  • significant unexpected signs
  • RESET test
3
Q

Omitted Relevant Variable

- remedy?

A

Include omitted variable or a proxy

4
Q

Irrelevant Variable

- what is it and what causes it?

A

The inclusion of an unnecessary variable

5
Q

Irrelevant Variable

- what are the consequences?

A

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

6
Q

Irrelevant Variable

- how to detect it?

A
  • theory
  • t-test on beta
  • adjusted r^2 increases if variable is dropped
7
Q

Irrelevant Variable

- remedy?

A

Exclude the irrelevant variable

8
Q

Incorrect Functional Form

- what is it and what causes it?

A

The functional form of the model might not be linear

9
Q

Incorrect Functional Form

- what are the consequences?

A
  • biased and inconsistent estimates

* poor fit of model (low R^2)

10
Q

Incorrect Functional Form

- how to detect it?

A
  • theory
  • Ramsey RESET
  • scatter plot of Y with each of the X’s
11
Q

Incorrect Functional Form

- remedy?

A
  • transform data into logs to linearise model

* add higher order functions of the variables to capture curvature

12
Q

Multicollinearity

- what is it and what causes it?

A

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

13
Q

Multicollinearity

- what are the consequences?

A
  • high R^2, coefficients high SEs -> low t-ratios
  • regression sensitive to small changes
  • wide confidence intervals for parameters, incorrect inferences from model
14
Q

Multicollinearity

- how to detect it?

A
  • Correlogram

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

15
Q

Multicollinearity

- remedy?

A
  • ignore if model is okay
  • drop collinear variable
  • transform correlated variables into a ratio
  • collect more data (longer sample period, higher frequency obs)
16
Q

Autocorrelation

- what is it and what causes it?

A

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

17
Q

Autocorrelation

- what are the consequences?

A
  • unbiased but inefficient
  • incorrect inferences
  • inflated R^2
18
Q

Autocorrelation

- how to detect it?

A
  • Durbin Watson
  • Breusch Godfrey
  • Correlogram of residuals
  • Ljung-box
19
Q

Autocorrelation

- remedy?

A
  • GLS (if form is known)
  • Dynamic Models
  • HAC coefficients
  • SE Newey-West
20
Q

Heteroscedasticity

- what is it and what causes it?

A
Variance of error term not constant for all observations
Causes: • scale/size effects
• measurement error
• subpopulation differences
• flow of info is time varying
21
Q

Heteroscedasticity

- what are the consequences?

A
  • unbiased but inefficient estimates

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

22
Q

Heteroscedasticity

- how to detect it?

A
  • visual inspection of residual plot graph
  • white’s test
  • engle’s LM test for ARCH
23
Q

Heteroscedasticity

- remedy?

A
  • GLS (if form is known)
  • transform variables using logs
  • white’s SE estimates
24
Q

Seasonality

- what is it and what causes it?

A
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
25
Q

Seasonality

- what are the consequences?

A

Serially correlated error

26
Q

Seasonality

- how to detect it?

A

Dummy variable for the period where the pattern is observed

27
Q

Seasonality

- remedy?

A

Intercept or slope dummy to account for seasonality

28
Q

Normality

- what is it and what causes it?

A

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

29
Q

Normality

- what are the consequences?

A
  • test statistics do not follow normal distribution
  • estimators not efficient
  • SEs are biased leading to wrong inferences
30
Q

Normality

- how to detect it?

A
  • Bera-Jarque
  • histogram of residuals
  • skewness and jurros is
31
Q

Normality

- remedy?

A
  • dummy variable to knock out outliers

* GARCH model

32
Q

Structural break

- what is it and what causes it?

A

Parameters are not constant over sample period

33
Q

Structural break

- what are the consequences?

A

Biased coefficient estimates

34
Q

Structural break

- how to detect it?

A

• chow test for structural break

35
Q

Structural break

- remedy?

A
  • split the period

* dummy variables to account for different behaviour over the periods