QTM 100 Study Guide - Quiz Guide: Type I And Type Ii Errors, Mann–Whitney U Test, F-Distribution

• School: Emory University
• Department: Quantitative Theory & Methods
• Course: QTM 100
• Professor: Jennifer Mc Gee

Document Summary

Created by latha karne ox 17 c 19. Each variable is linearly related to the outcome. Same as linear regression, but many predictor variables. Slope: for every unit increase in x, the y is expected to increase by (whatever the slope is), given the other variables are held constant. Line of best fit: tells us y will increase by slop e 1 for every unit increase in x 1 and slop e 2 for every unit increase in x 2. Increases only if the new term improves the model more than would be by chance. Decreases when a predictor improve the model less than explained. Problem 2: model has too many predictors and higher order polynomials, begins to model the random noise in the data overfitting the model and gives a higher r 2. *question that could come: given a scenario and asked whether it would produce a higher r 2 than normal or a smaller one.