Chapter 4 (4.6-4.7 + 4.9) - Predicting Outcomes Using Linear Regression Flashcards
R^2 is an indicator of what?
The coefficient of determination.
What does the coefficient of determination range from?
0 to 1.
What does the coefficient of determination measure?
The proportion of the variation of Y explained by the model; the higher the coefficient of determination, the better the model fits the data.
What does “SSR” stand for?
The sum of the squared residuals; measures the variation of Y not explained by the model.
What does “TSS” stand for?
The total sum of squares; measures the total variation of Y, explained and unexplained.
True or false: the coefficient of determination should only be compared between models that have the same outcome variable.
True
How can the slope coefficient be interpreted?
As the expected change in the predicted Y when X increases by 1.
What does R^2 measure?
The proportion of the variation in the outcome variable explained by the model.
True or false: the lower R^2 is, the better the model fits the data.
False
Account for the R^2 simple linear model.
R^2 = cor(X,Y)^2
What does it mean when the correlation is either -1 or 1?
That the relationship between X and Y is completely linear.
What does it mean when the correlation is 0?
That the relationship between X and Y is non-linear.
