Data science is a very open field. Nearly everything we use is contributed by the community. Yet, we don't share our failures with the community for some reason. I think we learn more from failure than from success, so I decided to share how I seriously messed up a recent project.

I thought it would be great to build a forecasting model for the US housing market. HA! I made several big mistakes: - I started with a vague idea. - I didn't know if the data was available or easy to get. - Failing to start over and reassess when I realized the project was going in weird directions. - The model's objective wasn't directly connected to the way the model might be used. - I failed to appreciate the core of the problem was effectively predicting the economy, which is a much bigger and more complex problem.

I might not always make a blog post, but I will keep sharing my failures. Because we shouldn't be crabs in a bucket.