Section 27.8 Section 5.4 Summary
You continued to study least squares regression in this section, turning to issues of inference. You learned the conditions of the basic regression model (linearity, independence, normality, and equal variance), and once again you used simulation to study the approximate sampling distribution assuming the null hypothesis is true, this time for sample regression coefficients. You found that the t-distribution can be used to perform significance tests and construct confidence intervals for regression coefficient parameters and for predicted values, based on sample data. The primary significance test of interest is usually whether the sample data provide evidence of an association in the population between the two variables, which is equivalent to testing whether the population slope coefficient equals zero. You also learned how to use residual plots to check the technical conditions associated with the basic regression model. Example 5.3 illustrates an application of regression analysis.
You have attempted of activities on this page.
