Intuitive Portfolio Shrinkage That Even Your Mum Can Understand
For all of you who are degenerates at constructing portfolios
How to lose money trading
Hi, I'm quantymacro. A hedge fund quantitative macro trader + researcher.With a lot of luck, I self-taught myself into receiving multiple offers from PMs in Citadel/MLP/Bluecrest/Brevan to join their pod without going through the typical grad rotations out of undergrad (my non-guide on how I did it).
I spend a lot of hours studying, reading books and papers about machine learning, quantitative finance, and portfolio optimisation, so you don't have to (but you still should). Subscribe Read by practitioners from top firms, including;
Like most aspiring quants, I was also a victim of poorly curated information out there. I wish I can get back the time I spent studying cointegration, hurst exponent, regressing prices etc. This blog is my *small* attempt to improve the signal-to-noise ratio of content out there.
Trading stuff that amateurs traders seem to care about a lot, and which are mostly ignored or seen as trivial by practitioners -
— cephalopod (@macrocephalopod) May 26, 2022
1. Stationarity / unit root tests for time series
2. Cointegration
3. Volatility forecasting
4. Exit signals
5. Stop losses
Honest-to-God Reviews
...has all the qualities to potentially be a good trader. if only he reads some news about the Fed instead of just reading some fake maths.
Senior Macro PM at Millennium
...very pleasant as a lunch buddy. can perform some card magic tricks as well surprisingly.
Strategist at Citadel
...love him a lot but he should get a real job with real skills.
Mom
Comments from good people, not necessarily because my writing is good
Little project: In the next few days, I'd like to link to/comment on a few posts (long-form, on the web) by @KrisAbdelmessih, @quantymacro and @markku_kurtti, and @0xfdf. For the simple reason that they are good, I am (re-)reading them and they should not be forgotten. I was…
— Giuseppe Paleologo (aka gappy) (@__paleologo) July 29, 2024
@__paleologo
ex-Director of Quant at Citadel, Head of Risk Enterprise at Millennium, Head of Risk Management at Hudson River Trading
I read this, it's good, I guess my only comment would be that I would be careful about taking my opinions as facts because I know many successful PMs who do things completely differently to me and it clearly works for them 😅
— cephalopod (@macrocephalopod) March 14, 2024
@macrocephalopod
X pseudonym
i really liked your ridge regression thing you sent today 🙏
— Robot James 🤖🏖 (@therobotjames) April 4, 2024
@therobotjames
X pseudonymPhilosophy of quantymacro's Blog
This blog is not focused on backtesting random entry/exit rules, or Python snippets on how to fit complex ML models on price data. The blog focuses heavily on understanding the inner workings of quantitative methods, as well as some pitfalls that the author has fallen into. Yes, sometimes this does mean that we will go back to the basics (you probably need it). My biggest nightmare is going into a quant interview with ML in my CV and bomb a linear regression question.
I'm hiring a junior quant and out of 30 applicants only 4 have passed my first round interview. Despite having tier 1 math/stats degrees and prior roles, most are weeded out if you simply ask them basic questions about regression. I'm talking chapter 1 linear model stuff here. https://t.co/6yhAlbyAe6
— fdf (@0xfdf) May 7, 2024
the growing popularity of machine learning had a large negative impact on the average quality of applied finance projects in the MFE program I taught in -- we would warn teams extensively of the pitfalls but still 30% of them turned into exactly this https://t.co/r8CYH4yckv
— Benn Eifert 🥷🏴☠️ (@bennpeifert) March 30, 2022
It’s a strong opinion of mine that really intuitively understanding the properties of time series filters and really understanding the properties of univariate and multivariate OLS will pay off many times over.
— cephalopod (@macrocephalopod) December 5, 2023
Check out selected topics
More issues
Lowkey-Advanced Ridge Regression (Part II): Non-Zero Priors, Subset Shrinkage, Cluster Shrinkage
I Studied Regressions *Only* for 30 Days So You Don't Have To But You Have To Subscribe Part II
Lowkey-Advanced Ridge Regression (Part I)
I Studied Regressions *Only* for 30 Days So You Don't Have To But You Have To Subscribe
A Non-Guide on Building Personal Projects For HF Quant Roles (Part I)
yes I finally wrote about it.
Farewell Episode: Risk Parity Without Correlations
Never get into a committed relationship with assets' covariance. They're too unstable.
Cross-Predictability Between Assets (Part I): All Signals Matter
As the King of the quantymacro Land, I have passed an executive order of "No Signal Left Behind" effective immediately.
All my homies hate LASSO
I fell out of love with LASSO. it's not you it's me baby :(
the start of my grifting era.
yes I'm starting a "newsletter".
Should you combine positions or signals? (Part I): Having your cake and eating it too
Let’s just assume we have 2 signals (Trend and Carry). How do we combine these signals together? Generally there are two broad ways of doing it:
1. Combining positions of the individual signals
2. Regress future returns against the signals to get expected returns of each asset
In this
Should you combine positions or signals? (Part II): Having your cake and eating it too
Do we actually have to extract assets' expected returns to harvest the benefits of factor model? Here we will talk about another approach where we can sum up our positions, and still integrate a factor risk model, without having to explicitly extract the expected returns of the underlying assets.
subscribe to quantymacro
sometimes I write cool stuffs
email
Subscribe