My proyect in US Medical Insurance Costs

Here is my proyect on US Medical Insurance Costs. Having finished the Analyse Data with Python path I used the following libraries to complete this proyect.
Numpy
Pandas
Matplotlib
Seaborn

Love to get feedback form everyone.

3 Likes

Congratulations on finishing up, you seem to have reached some reasonable conclusions from your analysis. For a bit of feedback most of my suggestions are on the presentation of data.

For example this is an except form part of the output which would mean very little to anyone who hasn’t used the pandas library-

 5   region    1338 non-null   object 
 6   charges   1338 non-null   float64
dtypes: float64(2), int64(2), object(3)
memory usage: 57.6+ KB

Perhaps there’s a better to display that information (if it’s provided in other tables or figures then is it worth including?).

I’d also advocate for formatting some of your values, mainly the floats, as the inclusion of multiple decimal places isn’t actually useful for getting your message across. Average age of 40.4 is easy for a human to understand, 40.423058534 is not as easy.

There’s at least one table one smoking which has no description and isn’t mentioned in the text so it’s not entirely clear what it’s for. Double check that all tables and figures are mentioned in the text (ideally with some sort of labelling system e.g. table2 or figure 10) and try and make them as easy to read as possible (ideally you could understand them without ever reading the text).

You make a decent comment about increasing insurance costs due to smoking (and mention a rough increase of 3 times) but perhaps you could back that claim up with the data (ideally with errors) for example “an increase of 3.1 +/- 0.2” or whatever useful statistic you have derived and if possible a similar statistic for age or at least refer to a figure or table where you analysed that data.

The work is all there (analysis and coding), but it is worth considering how easy it is for someone else to understand the information you have obtained. Presentation of data counts for a lot.

Thank you for taking the time to see my code. I would have used Medium tu post the work but I wanted to move foward in the course. Anyway, I will take your advise for future presentations.