search

ijyliu / ECMA-31330-Project

Econometrics and Machine Learning Group Project
2 stars 1 forks source link

Restructure Code to Emulate PS1 Question 3 Structure #16

Closed ijyliu closed 3 years ago

ijyliu commented 3 years ago

Implement something similar to https://github.com/ijyliu/ECMA-31330-Project/blob/main/Misc/PS1_Q3.ipynb ie, create a multi-index of N, rhos, any other parameters we want to vary, and apply a simulation function to it.

Make a dataframe of the cartesian product of simulation parameters

No need to store the actual data; instead, just store the true betas, the measurement error, the IV and PCR coefficients

ijyliu commented 3 years ago

Actually, there are probably even better ways to structure this, particularly if things will be run in parallel.

I guess the focus should be on writing functions, which can be easily moved around from file to file as needed...

ijyliu commented 3 years ago

Current code does this very roughly... but it seems to work.

ijyliu commented 3 years ago

This issue is more or less complete