Exam AWS Certified Machine Learning - Specialty topic 1 question 96 discussion

I would choose A. See: https://www.itl.nist.gov/div898/handbook/pmd/section4/pmd442.htm and https://blog.minitab.com/blog/the-statistics-game/checking-the-assumption-of-constant-variance-in-regression-analyses

Ans. is A. High-degree polynomial transformation.

Answer A. One of the key assumptions of linear regression is that the residuals have constant variance at every level of the predictor variable(s). If this assumption is not met, the residuals are said to suffer from heteroscedasticity. When this occurs, the estimates for the model coefficients become unreliable https://www.statology.org/constant-variance-assumption/

Agree with A

https://blog.minitab.com/en/the-statistics-game/checking-the-assumption-of-constant-variance-in-regression-analyses

Kind of like heteroskedasticity, anyways it is A.

Answer is A. It does not has constant variance !

A is correct answer

These images are broken. I cannot review the question properly!

D is best answer all x values are scattering as a whole , no matter what x is https://www.statisticshowto.com/residual-plot/ if you take all x values to plot histogram , it will be bel-curv.

And it does NOT mean linear regression is not appropriate. It means your linear regression model is biased due to several reasons.

100% A

I will choose A , because the data is heteroscedastic. It violates a key assumption of linear regeression

A. https://www.originlab.com/doc/origin-help/residual-plot-analysis

Do not have content variance https://stats.stackexchange.com/questions/52089/what-does-having-constant-variance-in-a-linear-regression-model-mean

Answer is A. As x raises, the residuals become higher and higher...

Some Good Reading https://www.andrew.cmu.edu/user/achoulde/94842/homework/regression_diagnostics.html Ans is A