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;
30 Cards in this Set
- Front
- Back
log-likelihood function
|
the sum of the log-likelihoods, where the log-likelihood for each observation is the log of the density of the dependent variable given the explanatory variables; the log-likelihood function is viewed as a function of the parameters to be estimated
|
|
beta coefficient or standardized coefficient
|
regression coefficient that measures the standard deviation change in the dependent variable given a one standard deviation increase in an independent variable
|
|
Goldfeld-Quandt test
|
assumes that you know the form of the regression so that you know how to partition the data. But, gives an exact test for even small samples.
|
|
variance of the prediction error
|
the variance in the error that arises when predicting a future value of the dependent variable based on an estimated multiple regression equation
|
|
maximum likelihood estimate (mle)
|
a broadly applicable estimation method where the parameter estimates are chosen to maximize the log-likelihood function
|
|
dummy variable trap
|
the mistake of including too many dummy variables (variables that take on the value of 0 or 1) among the independent variables. Occurs when an overall intercept is in the model and a dummy variable is included FOR EACH GROUP.
|
|
Breusch-Pagan test
|
a test for heteroskedasticity where the squared OLS residuals are regressed on the explanatory variables in the model
|
|
asymptotic efficient
|
for consistent estimators with asymptotically normal distributions, the estimator with the smallest asymptotic variance
|
|
LaGrange-Multiplier Test (aka score test)
|
used to estimate the improvement in model fit if additional variables were included in the model, an test for omitted variables.
|
|
central limit theorem (CLT)
|
a key result from probability theory that implies that the sum of independent random variables, or even weakly dependent random variables, when standardized by its standard deviation, has a distribution that tends to standard normal as the sample size grows
|
|
adjusted R-square
|
a goodness-of-fit measure in multiple regression analysis that penalizes additional explanatory variables by using a degrees of freedom adjustment in estimating the error variance
|
|
oblique v. orthogonal projections
|
oblique: graphical projection using 2-dimensional depictions of 3-d objects
orthogonal: maps it in 3-d space (so oblique matrices have only (x,y); orthogonal have (x,y,z) |
|
method of moment estimators
|
an estimator obtained by using the sample analog of population moments; OLS and two-stage least squares are both method of moments estimators
|
|
probit model (of being married)
|
a model for binary responses where the response probability is the standard normal cdf evaluated at a linear function of the explanatory variables
|
|
law of large numbers
|
theorem: the average from a random sample converges in probability to the population average; also holds for stationary and weakly dependent time series
|
|
variance inflation factor
|
in multiple regression analysis under the Gauss-Markov assumptions, the term in the sampling variance affected by correlation among the explanatory variables
|
|
prediction interval for forecast
|
a confidence interval for an unknown outcome on a dependent variable in a multiple regression model
|
|
instrumental variables
|
in an equation with an endogenous explanatory variable, an IV is an ommitted variable, is uncorrelated with the error in the equation, and is partially correlated with the endogenous explanatory variable
|
|
two-sided alternative
|
you're saying that the independent variable WILL have an effect, but you're not saying in which direction. It could be positively or negatively correlated.
|
|
exclusion restrictions
|
restrictions that state that certain variables are excluded from the model (or have zero population coefficients)
|
|
Breusch-Godfrey test for AR(q)
|
an asymptotically justified test for AR(p) serial correlation, with AR(1) being the most popular; the test allows for lagged dependent variables as well as other regressors that are not strictly exogenous
|
|
pure v. impure heteroskedasticity (Butler's notes)
|
Pure- there are no correlated omitted variables that cause the variance to change (o2=o2zi2)
Impure- The omitted variable is correlated with the independent variable (o2=f(zi,zi)) |
|
"robust" option in STATA
|
a STATA code that keeps the OLS coefficientes but uses robust standard errors.
|
|
"prais y x1 x2 x3" procedure in STATA
|
stata junk
|
|
between-group v. within-group estimators
|
dunno about the estimator part...but you basically want your between-group differences to be much higher than within-group for your independent variable. Like if I was doing a study where I changed your blood to gasoline, I'd want there to be more variance between your blood and the control group's than between your blood and the other sap with gas in his veins. Get it?
|
|
dprobit in STATA
|
stata crap
|
|
marginal effect from a logit regression
|
dP(X)/dXj = g(B0+XB)Bj where g(z) = dG/dz (z)
|
|
STATA "test" statements for this model: regress y x1 x2 x3; test (x1=x2) (x3=x0)
|
stata stratosphere
|
|
weighted least squares estimator
|
an estimator used to adjust for a known form of heteroskedasticity, where each squared residual is weighted by the inverse of the estimated variance of the error
|
|
"hettest, rhs" in STATA
|
tests for heteroskedasticity in gression model for a right hand sided test.
|