I am often asked: how can I study more effectively for your class? How
can I understand the material better?
I don't have a good answer. Effective learning techniques vary from
person to person.
Here are some things that worked for me:
Read a lot. I tried to get my hands on several texts for each
model covered in class. You only realize what you did not understand
when you see it presented in different ways.
Write. I rewrote all the class notes for two reasons:
It keeps things organized. (Back in the day we did not have
transparencies that could be distributed as pdf files.)
It forces you to go over each step and fill in the missing
details.
More obvious tips:
Do lots of problems. Don't look at the answers until you get
stuck.
Ask questions early.
Working With Data
Often the easiest way to make sure everything is correct is to create a
test dataset. That way all the expected results your code should
generate are known.
Write your code so that you can go from raw data to final results
without any manual intervention. You should be able to erase everything
you have computed and recreate it on autopilot.