Linear Regression Flashcards
When do we use linear regression?
Predicting the outcome of 1 numeric variable based on the values of another numeric variable
What is the equation for linear regression?
y^ = a + bx
Interpretation for linear regression
__% of the variability in ____ (y^) is accounted for by the variability in ____(x)
Assumptions for linear regression
Relationship between x and y is linear (use residual plots to find no curvature), Errors are normally distributed (histogram and box plot for symmetry, outliers), Errors have constant variance for all x-values (look at residual plot of vertical scatter being the same)
Hypotheses
H0: ____ is not a linear predictor of ____
Ha: ____ is a linear predictor of _____
What values do we list?
Test statistic (t or F), p-value
Conclusion
There is/isn’t sufficient evidence of a linear relationship between _____ and _____
CI
We are 95% confident that as ____ increases by 1 ___(units) ______ increases/decreases between _____ and ____ on average
How do you find residual values
y - y^