Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
18 Cards in this Set
- Front
- Back
What is assumption I and what are its violations?
|
Dependent variable is a linear function of a specific set of independent variables plus a disturbance term.
Y=Xß+Σ Violations: Wrong regressors Nonlinearity Changing parameters |
|
What is assumption II and what are its violations?
|
Expected value of a disturbance term is zero.
EΣ=0 Violations: Biased Intercept |
|
What is assumption III and what are its violations?
|
Disturbances have a uniform variance and are uncorrelated
EΣΣ'=σ²I Violations: Heteroskedasticity Autocorrelated Errors |
|
What is assumption IV and what are its violations?
|
Observations on independent variables can be considered fixed in repeated samples.
X is fixed in repeated samples Violations: Errors in variables Autoregression Simultaneous equations |
|
What is assumption V and what are its violations?
|
No exact linear relationships between independent variables and more observations than independent variables.
Rank of X=K≤T Violations Perfect multicollinearity |
|
What is wrong regressors and what are the consequences?
|
Violation of assumption I: when you include regressors that should not be included in the model, or you do not include regressors that should be in the model
Consequences: Omitted variable bias Parameter inconsistency |
|
What is nonlinearity and what are the consequences?
|
A violation of assumption I: When the true relationship between the dependent variable and independent variables is non-linear.
Consequences: Biased parameter estimates Inflated standard errors Lower R Square |
|
What are changing parameters and what are the consequences?
|
Violation of assumption I: any time your parameters do not stay constant during the time period in which they were collected
Consequences: Biased parameter estimates Inflated standard errors Lower R square |
|
What is a biased intercept and what are the consequences?
|
Violation of assumption II: when the expected value of your error term is not zero.
Consequences: Biased intercept Biased parameters |
|
What is heteroskedasticity and what are the consequences?
|
Violation of assumption III: When the shape of the disturbance term is not constant
Consequences: Inefficient parameter estimates Loss of precision |
|
What are autocorrelated errors and what are the consequences?
|
Violation of assumption III: disturbances are correlated, usually over time
Consequences: Loss of precision |
|
What are errors in variables and what are the consequences?
|
Violation of assumption IV: variables are incorrect because they are measured incorrectly from one sample to the next
Consequences: Biased parameters Inconsistent parameters |
|
What is autoregression and what are the consequences?
|
Violation of assumption IV: using the value of the lagged dependent varaible as an independent variable
Consequences: Biased parameters Inconsistent parameters |
|
What are simultaneous equations and what are the consequences?
|
Violation of assumption IV:
phenomena determined by simultaneous interaction of several variables |
|
What is perfect multicollinearity and what are the consequences?
|
Violation of assumption V:
variables are linearly related to each other Consequences: Unstable parameter estimates Loss of precision |
|
Cross Sectional Data
|
Multiple economic units for the same time period. (One observation only)
Example: asking about individuals' voting preference for this year, one time only. |
|
Times Series Data
|
Uses multiple observations with the same economic unit over time
Example: asking one student whether they voted Republican or Democrat in each of the past presidential elections. |
|
Panel Data
|
Cross sectional time series data (multiple observations on more than one economic unit over time)
Example: asking multiple students whether they voted Republican or Democrat in each of the past presidential elections |