ECON10005 Chapter Notes - Chapter 3: Standard Error, Regression Analysis, Variance
Document Summary
Regression models predict the value of one variable on the basis of another. For any variables (cid:1851) and (cid:1850), the linear regression model is. : error term / disturbance (cannot find (=deviation of (cid:1851) from its population (cid:2181): dependent variable (cid:2180): explanatory variable / regressor. , : estimators of parameter and . Best fit line / sample (estimated) linear regression model because population mean is unknown) The line that has best fit to the data sxy: sample covariance conditional mean) (independent) = (cid:1799) (cid:1799) = (cid:1799) ( (cid:1798)) (cid:1815) (cid:2778) (cid:1793)(cid:1793) = (cid:2779) Best fit: , sse min (sum of. Residuals capture the part of y not explained by x: 1. I: residual / error y i: fitted value yi: observed value. The estimated average testscore of class size with 20 students is. To test if the data fits the linear model, examine. Standard error of estimate (2) statistics. Hypothesis test (3) slope, intercept, conditional mean, correlation.