What assumptions do you need to be fulfilled when you are using regression?

What assumptions do you need to be fulfilled when you are using regression?

Assumptions in Regression

• There should be a linear and additive relationship between dependent (response) variable and independent (predictor) variable(s).
• There should be no correlation between the residual (error) terms.
• The independent variables should not be correlated.
• The error terms must have constant variance.

What transformations can be used when assumptions are violated in regression?

Possible alternatives if your data violate regression assumptions

• Different linear model: fitting a linear model with additional X variable(s)
• Nonlinear model: fitting a nonlinear model when the linear model is inappropriate.

How are transforming variables used to correct assumption failures?

Transforming variables can be done to correct for outliers and assumption failures (normality, linearity, and homoscedasticity/homogeneity); however, interpretation is then limited to the transformed scores. Normality assumes that the dependent variables are normally distributed (symmetrical bell shaped) for each group.

When to use estimated regression models based on transformed data?

Understand when transforming predictor variables might help and when transforming the response variable might help (or when it might be necessary to do both). Use estimated regression models based on transformed data to answer various research questions.

How can you tell if the assumption of linear regression is met?

The easiest way to detect if this assumption is met is to create a scatter plot of x vs. y. This allows you to visually see if there is a linear relationship between the two variables.

When to use transformation or weighted least squares regression?

If there are unequal error variances, try transforming the response and/or predictor variables or use ” weighted least squares regression ” (see Lesson 10).