Regression

• regression is the nature of the relationship
• regression line 


  line that best fits the data
• regression equation

  y=a+bx
• y=a+bx

  a is the intercept
• y=a+bx

  b is the slope
• residual error is the distance a point is from the regression line
• residuals 


  the distance between the observed value of the dependent and the predicted value
• fitted values are the values that lie on the regression line
• independent are the same as the explanatory
• dependent is the same the response
• independent

  one you change/control
• dependent

  the one you measure
• the y variable is the dependent
• the x variable is the independent
• the dependent is explained by the independent
• you must have a direction to do regression
• Residual plots

  A) predicted and observed values

  B) equal variance

  C) normality

  D) outliers

  
• the F value if the residual value
• the F value reported with the degrees of freedom
• F value will have two numbers for the degrees of freedom
• F value of 24 with degrees of freedom of 1 and 30 would be reported as

  F(1,30)=24