How I won my 14th bronze medal by using statsmodels on Kaggle’s Ames House Price competition

This morning I was very pleasantly surprised to wake up, clear the notifications on my phone, ans see that I won a bronze medal on Kaggle’s Ames House Price competition:-

I won the medal because I had incorporated Kaggle’s statistics library, statsmodels into the Jupyter Notebook that I had prepared.

I thought it would be a good idea, therefore, to show some of the statsmodels functionality that I employed to win the medal.

The quantile quantile plot (qqplot) is a graphical technique for determining if two datasets come from populations with a common distribution. A qqplot is a plot of the quantiles of the first dataset against the second dataset. In the example below, I performed a qqplot analysis on the target, which was the sale price of the homes listed in the dataset:-

The qqplot 2 samples is a variation of the qqplot method, but unfortunately I was unable to find much about…