Exploring Ebay Car Sale

I would appreciate if you can find out time to look into my work and criticize it.

https://app.dataquest.io/m/294/guided-project%3A-exploring-ebay-car-sales-data/9/next-steps

Basics.ipynb (167.8 KB)

Click here to view the jupyter notebook file in a new tab

4 Likes

Hi @aniefiokuduakobong ,

Your project looks great, the data analysis is very thorough and profound.
Here are some suggestions for your consideration.

  • Emphasizing more the project title. I just mean to make the letters bigger :blush: The same about conclusion.

  • In case of this dataset, it’s better not to remove all the prices below 25 and above 75 percentiles. They are all real prices (apart from the absurd maximum price that is clearly wrong), and these rows contain also a lot of other valuable information.

  • The markdown cell after the code cell [12] (and in some other parts of the project). Probably it would be better not to describe in detail the methods of data analysis that are going to be applied. In markdown cells it’s better to give some explanations of the results / observations rather than technical approaches.

For the rest, great job!
Good luck with your next Dataquest projects!

2 Likes

Good to see you again with another guided project. You are doing an awesome job by finishing tougher and tougher guided project one after the other! Great job!

So let me point out a few thing I would have done differently.

  • At the beginning you have printed the whole auto df. It would have been better to use the df.head() so that it can avoid some scrolling through the entire dataframe for the users/readers.

  • In order to rename the columns you have used a function and a loop. But you can also achieve the same by using

dataframe.rename({'oldname':'new_name'},axis=1, inplace=True)

This way it saves a few lines of codes. Also, you have used this same format later to rename the price and odometer column. So why not for the whole column names as well? Maybe you want to experiment with other techniques which is nice.

Its good to see that you were playing around with outliers. Since I didn’t know much about Outliers while doing this project, I didn’t remove the outliers as such. But I was removing rows that didn’t make much logical sense such as price being 0, registration year beein 1000 and 9999. I’m not sure if finding outlier is the right approach in those cases.

While doing the project, I also felt that there could be more than one reasons for some high values or seemingly wrong values. What if someone is actually selling a vintage car made in 1908 which then can have a very huge price tag ( In some rare case this might be a possibility). So I felt a deeper inspection was needed before removing any suspicious values.

Also like @Elena_Kosourova said, you can highlight your conclusion part because mostly that is what many of the end users are interested in.

But going through your project was a pleasure. Everything is well explained, used very generous amount of inline codes to explain. So it was a great read.

If you want to have a look at what I have done and send me a feedback, here it is.

Hope this was helpful. Happy learning.

1 Like

@Elena_Kosourova. Thanks for your positive criticism to make me improve more. it means alot to me. i will work on the parts you have mentioned and will send the modified copy to you for review. once again tahnks. :raised_hands:t2:

1 Like

it really was helpful. @jithins123 thanks for stopping by to comment on my project work. As you have rightly stated, i will rework on the parts you have highlighted and tag you to see the modified copy.
Thanks again. i appreciate :raised_hands:t2:

1 Like

i have just gone through your project. Hmmm! i still have a very long way to go.
your work us very detailed and well explanatory. welldone

Hi, hope you are doing good.
Glad it was helpful. Looking forward to going through it again and thank you for going through my project as well.