Week 7: Linear Regression Models Flashcards
Correlation and regression analysis help us to
Best evaluate relationship among variables
Regression analysis
We believe one or more variables may influence the variation of another variable
Correlation
We are interested to evaluate the relationships among variables without specifying roles for variables
A scatter plot gives information about two characteristics of a relationship
1) direction, 2) form and 3) strenght
The direction of a relationship may be
Positive, negative or not clear
The form of a relationship in a scatter plot might be
1) linear 2) non linear
A linear relationship may be
1) present or 2) absent
The strength of a relationship might be
1) strong or 2) weak
When does a scatter plot shows no relationship between variables?
When the data follow a flat straight line with the slope equal to 0
What is the goodness of fit?
When we determine how well the response is explained by the explanatory variable
What is the goodness of fit?
When we determine how well the response is explained by the explanatory variable
The closer the points are to the line, the better the regression line fits the data
True or false
True
How do we call the distances between the points and regression line ?
Residuals
How do we symbolize the coefficient of determination
R2
As a percentage R2 is a value between 0% and 100%
True or false
True
When do we say there’s a high goodness level?
When the points of the plot are close to the regression line
When do we say there’s a los goodness of fitness?
When the points are farther to the regression line
Which kind of test can we use to determine wether the explanatory variable is related to the response variable?
Hypothesis test on the slope
