TOPIC 5 MULTI REG Flashcards
When using this model, we assume that it is….
1. Linear in the parameters
2. Nonlinear in the parameters
- Linear in the parameters
When using this model, we assume that
1. Fixed X values are dependent of the error term
2. Fixed X values are independent of the error term
- Fixed X values are independent of the error term
Independent vars (X) that are correlated with the error term are called…
ENDOGENOUS
a situation where an explanatory variable (independent variable) is correlated with the error term in a regression model. This correlation can lead to biased and inconsistent estimates if ordinary least squares (OLS) is used.
What is the mean value of the disturbance ui
= 0
What does homoscedasticity mean?
the spread of the residuals does not change
regardless of the value of the predictor variables.
We must assume that n must be (greater/less than) the # of parameters to be estimated
GREATER
Should there be a variation in the values of the X vars? Why or why not?
YES, if there is no var then we cannot explain our regression model properly
No ____ collinearity between the X vars
EXACT
Is there ever collinearity or multicollinearity?
NEVER
What does no collinearity mean?
None of the regressors can be written as exact linear combos of the remaining regressors in the model
The multi reg equation gives…
the conditional/expected mean of Y conditional upon the given/ values of X1 & X2
an equation to get (E)Yi
What does a high error variance cause?
An increase in the sampling variance because there is more “noise” in the equation
What does a large error variance lead to?
Leads to imprecise estimates
Does the error variance have a parallel relationship with sample size?
No, when sample size decreases error var wil not also decrease
Should you increase sample variation?
Yes
It leads to more precise estimates (ceteris paribus)
