Econ 2p91 Review:R square lies between 0 and 1R square equals that 60 percent of the variation in the dependent variable is explained by variation in all the independent variables included in the model. R square has disadvantage that the value always increase as new independent variables are added to the model no matter they are relevant or not. This problem is solved by using adjusted R square.Report the regression results using the same format on the assignment: R square, standard error and SER. Coefficient of determination (r-square): apply units into formula. Coefficient of AGE: it means one unit change in AGE, there will be XXX change in AHE. Holding Female constant. 95% confidence interval: +-1.96 SESER = 11.5. It means that the model has high value of error. SER also indicates the level of goodness of fit. In this example, SER of 11.5 is a large number, it indicates poor fit.SER has the same unit as the unit of dependent variables (in this case, it is xxx)Multicollinearity is not a concern in this model, as we are dealing with only one regressor.
The confidence interval for the sample regression function slope: can be used to conduct a test about a hypothesized population regression function slope.The Gauss-Markov theorem states the OLS estimator is BLUE provided that all the standard assumptions are met. The U in BLUE means that: the longest.All the following statements refer to the Gauss-Markov theorem: the sum of OLS residuals is zero; OLS estimator is linear in the dependent variables; the mean of sampling distribution of the OLS estimator of the regression slope is equal to the value of the corresponding population parameter. The confidence interval for a single coefficient in multiple regression: should not be computed because there are other coefficients present in the model.Binary variables can only take on two values.How to determine direction of causation is to determine who is being x.A binary variable trap in econometrics is multicollinearity problem. OLS has the smallest variance.Under perfect multicollinearity, OLS estimator cannot be computed. When there are omitted variables in the regression, which are determinants of the dependent variable, then the OLS estimator is biased if the omitted variable is correlated with the included variable.