Final Report Review by zw429

[zwangzh]

This is an interesting report discussing an everyday issue: house buying. I really like the topic as this could be helpful to a lot of people. This report used data of house pricing and related features of NYC, and used models to predict the housing price and see whether the pricing of the house is reasonable for a buyer.

Things I liked:

• This is indeed an interesting topic, it would solve the common issue that a lot of people face.
• This report contains good amount of good descriptive statistics. And the data visualization really helps readers understanding the trend and story.
• This report is written based on good research of background information.
• This report provides good reasoning for using/not using certain methods of data analysis.
• This report is clearly formatted and written.

Things that could be better:

• When plotting the histograms of pricing, the 0s/missing data can be eliminated to show details of the normal ones. (Since the readers already know that there were lots of missing data.)
• Sales price and actual price and be visualized better to show a clearer comparison.
• Similarly for regression models, better to compare them side by side like using subplots.
• I would be good to provide some thoughts on why would l1 work the best.
• It would be nice to have better graphs than histograms for some of the visualization.

Overall, this is a good clearly written report about an interesting issue.

[AaronZang]

@zwangzh Thanks for the comment, especially those about how to polish the report! Appreciate it.