[QUESTION] Multiple regression assumptions not met - implications and solutions?
Hi everyone,
I’m currently working on my undergraduate thesis using multiple linear regression. After running the assumption tests, I found that 3 assumptions are not met:
1.Linearity
2.Normality
3.Homoscedasticity
Now I’m a bit confused about how serious this is for the validity of my analysis and what does this implied for my research, and the overall quality of the research.
Also, I’m still unsure:
1.How severe are these violations in practice for multiple regression?
Can the regression results still be interpreted if several assumptions fail simultaneously?
What are the best solutions or alternatives usually recommended in academic research
Some possible solutions I’ve read about hasn't been taught and really complex
Has anyone dealt with a similar situation in their thesis/research? What did your supervisor or examiner usually recommend?
Thanks a lot!