Typically, when doing such regressions you want to use changes, but the question is what type of changes? 1/n

https://x.com/j_fishback/status/1950750394977497117

We use regression models to do inflation forecasting, so it's a super important for us to understand how we preprocess the variables before feeding to the regression. 2/n

Great thread on backtesting errors in assets with a large number of names (eg. look ahead bias in picking your universe), inspired, I'll write about a few problems you can encounter with economic data backtesting 1/n

https://twitter.com/macrocephalopod/status/1598823745681903616

This thread is mostly inspired by the work we've been doing @turnleafan forecasting inflation with machine learning & alt data, which requires large amounts of data (including economic data) as an input.. & many months of cleaning/preprocessing it :-) 2/n

There've been lots of great threads on vol trading recently on fintwit from the likes of @VolQuant, @bennpeifert & many more! I thought I'd give it a go explaining how to backtest FX options + include Python code samples! 1/n @VolQuant @bennpeifert Before you get into creating a trading signal, need to create a total returns time series for your option instrument of choice. Say you enter a long 1m EURUSD ATM straddle, need to mark it at entry 2/n

a quick thread on EURUSD, given rally & dollar demise stuff (!), here's number of articles on dollar demise vs. spot.. historically peak together! from a historical perspective on PPP (CPI) spot is at fair value (undervalued on PPI though).. nowhere near all time high too

Been writing Jupyter notebooks for a Python, alt data/NLP + large datasets course. As a result been using many libraries for large datasets.. here are a few takeaways on large datasets in Python 1/8 If you've got a small time series dataset, Pandas is often easiest choice, but what if it doesn't fit in memory? Can batch calculations, but what about libraries who do this for you! 2/8