Given a data set of 100 observations (n=100) with one dependent variable ($y$) and 50 independent variables (k=50). Inference answers the following two questions:

1. Which subset of or combination of the 50 independent variables affect $y$?
2. If I were able to increase the value of $x_1$ by 10%, how much would $y$ increase?

Prediction answers a much simpler question:

1. Given new values for each independent variable, what's my best guess of $y$?