Life Expectancy and GDP Project (really grateful for your feedback)

Projects Portfolio Project: Life Expectancy and GDP
1

Hi guys,

Thanks a lot for interest in my LE and GDP Project. In this project, I have had the chance to work on many different types of analyses. I did branch out to build a Linear Regression to predict the LE based on the GDP data. Yet, having realized that the data isn’t representative of the larger population, I eventually gave up. Anyway, I hope you can spare some time to take a look at my project and give me some feedback.

Here is my project on Github:

Much appreciated!

2

Hey breyenguyen, really nice submission: I really like the high level approach you to took to finding the distributions to start off with and the detail you broke it down to.

One small suggestion with your code, when plotting a bunch of different metrics as subplots (or anything repetitious like really), that’s probably a good opportunity to make a loop of some sort (you can check out what I did in my code below, from my own submission for the project which you’re welcome to review for me btw!)

Screenshot 2021-08-04 at 11.15.45 pm
Screenshot 2021-08-04 at 11.15.45 pm2188×1382 327 KB

One another thing worth considering is the degree to which an increase in GDP impacted life expectancy (or vise versa). Does a 4x increase in GDP = a 4x increase in life expectancy? Does it depend on the country? What can that say about the causality between those two metrics, and are there other variables at play?

Cheers,
Tolly

1 Like
3

Hi Tolly,

Thanks a lot for your suggestions. Actually, after uploading my work onto Github, I did change the code a bit using loop to make it much simpler.

As for the degree to which an increase in GDP impacted life expectancy, I fitted a OLS Linear Regression model to quantify the impact and realized that I should build 2 different models: one for China and another one for the rest. I added polynomial terms to the model for China because the relationship between GDP and LE in China appears to be a curved line. However, I eventually got rid of the models as I thought there would be much more than the GDP and it would be better if I have a dataset with more data from other countries.

Anyway, thanks a lot for taking time to check and give me feedback. Cheers!

1 Like