top of page
A
B
C
D
E
F
G
H
I
J
K
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z

Assumptions

A short description of the assumptions underlying various statistical and data mining techniques, along with indicators of the severity of the consequences of violating each assumption. Associated courses explain each assumption in greater detail, provide information about how to check whether the assumption is violated, and explain what to do about it if it is.

Linear Regression

The assumptions of a linear regression are: linearity, homogeneity of variance, and normality of the error distribution.

Linearity
Linear regression requires the independent and dependent variables to have a linear relationship with each other. If the relationship between the variables is not linear, it makes no sense to run a linear regression.

High

Homogeneity of Variance (a.k.a. Homoscedasticity)
Linear regression also assumes that the variance of the errors for each value of the independent variable are independent (e.g., the variance of the error does not get larger as the values of the independent variable increase). If this assumption is violated, then the confidence interval is suspect.

Mid

Normality of the Error Distribution
Linear regression assumes the values of the dependent variable are normally distributed for every value of the independent variable. However, this assumption only effects the interpretability of the coefficients when the sample size is small.

Low

More Assumptions Coming Soon!

Please reload

Assumptions L

​FOLLOW DE

  • LinkedIn App Icon
  • scholarRound.PNG

© 2015 by Deirdre Kerr. Created with Wix.

bottom of page