click below
click below
Normal Size Small Size show me how
Regression
Terms
Term | Definition |
---|---|
Level of Confidence | 1) is the amount of Type I error implied by a test 2) 100 minus the level of significance. |
Steps in the t test | 1) Set up null and alternative hypotheses. 2) Choose a level of significance and therefore a critical t value. (d.f. = n-K-1) 3) Run regression and get an calculated t-score. 4) Compare t score with critical t value. |
Gauss-Markov Theorem | Given Classical Assumptions I through VI, the OLS estimator of βk, is the minimum variance estimator, from among the set of all linear unbiased estimators of βk, for k = 0, 1, . . ., K. That is, OLS is BLUE |
BLUE | BEST LINEAR UNBIASED ESTIMATOR |
Robustness | a goal of econometrics; the degree to which the regression model can produce the similarly qualitative results across multiple data sets |
Ordinary Least Squares | Regression technique that minimizes the sum of the square of the residuals (minimize the sum of squared difference between actual data points and the estimated line) |
Schoastic Error Term | added to regression equation to introduce all of the variation in the dependent variable that cannot be explained by the independent variables |
Residual | The difference between the observed value and the estimated function value. Difference between observed Y and estimated regression line |
Error Term | The deviation of the observed value from the (unobservable) true function value. Difference between observed Y and true regression line. |
T-test | Tests for the significance of explanatory variables. |
t-value | Coefficient/SE |
Critical t-value is calculated based on three items: | 1) Degrees of freedom = ? 2) One-sided or two-sided test? 3) Significance level (generally 5% or 1%)? |
Standard Error Depends on: | 1) Goodness of fit of equation (SEE, or Standard Error of the Equation) 2) Sample size – larger samples make SEE smaller 3) How widespread X’s are from their mean |
TSS | Total sum of squares (TSS) = Explained sum of squares (ESS) + Residual sum of squares (RSS). Squared variations of Y around its mean. |
ESS | Amount of the squared deviation of from its mean that is explained by the estimated regression line. |
RSS | Amount of the squared deviation of from its mean that is unexplained by the estimated regression line; OLS minimizes RSS |