PSYC 305 Lecture Notes - Lecture 19: Simple Linear Regression, Linear Regression, Regression Analysis
30 Mar 2017
School
Department
Course
Professor
Document Summary
Values of y are independent and are sampled at random from the population. The relationship between x and y is linear (linearity). Y is distributed normally at each value of x (normality). The variance of y at every value of x is the same (homogeneity of variances) These are the same assumptions as we have for anova, except for the linearity assumption. In simple linear regression, you only have one independent variable. In multiple linear regression, you have multiple independent variables. Residual means the difference between our predictions for y and our actual y. Standardized residuals are often used (residuals divided by their standard errors) This will give you the exact same information. If you have the kind of pattern that we see on the first table, this means that you have the violation of at least one assumption out of three. The linearity assumption is violated in this case.
