Simple Linear Regression Flashcards
What are y, x and u in the linear econometric model?
y is the dependent variable, values of x are the independent variables ( regressors), and u is the unobserved regression error.
What is the formula for the simple linear regression model?
y=B0+B1(x1)+u
What is the assumption of conditional mean independence?
E[u/x1]=E[u] meaning that the mean of the unobserved error for slices of the population does not depend on the value of x1.
What does E[y/x1] = B0+B1(x1) tell us?
That a one unit increase in x1 changes the expected value of y by B1, however y doesn’t always equal B0+B1(x1) for all units in the population.
How can the regression model ( y=B0+B1(x1) + u ) be rewritten?
y = E[y/x1] + u
What are fitted values?
Fitted values are values of y calculated by using the estimated coefficients B0^ and B1^. Fitted values are denoted with a ^. e.g. the models best guess at what y is given x1, based on the estimated relationship.
What are residuals and how do we calculate them?
Residuals (u^) measures how far off the prediction y^ is from the actual y. They are calculated by subtracting the fitted value (y^) from the actual value (y).
What is SSR and how do we calculate it?
SSR ( sum of squared residuals) is calculated using the equation. (yi-B0^ -B1^(xi1)^2 for all values from i=1 to n then added together.
What is OLS estimation?
Ordinary least squares (OLS) estimates that B0 and B1 are simply the choices of B0^ and B1^ that minimize the sum of squared residuals.
What is important to note about changes in the savings rate and population on economic growth.
Changes in the savings rate and population change only temporarily effect growth until steady state is reached again. This is because in steady state income per worker remains constant.
