I’ve been asked to write a thread on how retail traders can get started in Quant Trading, and what to do without an institutional setup. There’s a lot to learn, so please don’t complain later that I didn’t warn you! 🧵
People argue that hedge funds are stupid if they don't exceed a benchmark rate of return... I wanted to address that, so let's take a hypothetical crypto strategy, one that I've made with a 20% vol target. It looks worse than bitcoin, right? ... A 🧵
The brutal truth of finance is that name-brands matter.
My undergrad was at a non-target, and my quant career was a non starter.
After studying my masters at Cambridge - I got interviews at every top quant fund on the list. I hated that reality, but it is what it is.
I have received some questions either in DM or LinkedIn on universities and PhDs that can help you A LOT land a role in a top quant fund. I am probably not the best person to answer but here is what I think:
In Europe there are 6 universities that are considered targets pretty
I’ve been asked to write a thread on unorthodox feature engineering for quant strategies. It’s a bit of a long topic, so I appreciate your patience. I’ve tried to keep it simple and mathematically approachable.
I’ve been asked by my esteemed colleague
@goodalexander
to write a thread on recent industry trends within Quant Trading. My background is mostly within the mid-frequency space, so I’m likely to miss hugely obvious points to those in parallel industries. Apologies: 🧵
It is with great sadness that the Simons Foundation announces the death of its co-founder and chair emeritus, James Harris Simons. Jim was an award-winning mathematician, a legendary investor and a generous philanthropist.
Not to be that guy, but I seriously expect every single person on the sales/trading side of banking to be able to talk me through a vol surface.
If not, strong NGMI vibes.
I was asked to write a thread on the best HF coding practises and standards. I’m not a software engineer, I’m a mathematician who codes OK. I run many successful automated strats in a HF, and here are a few things I’ve learned 🧵
I was asked by an esteemed colleague,
@therobotjames
, to write a thread on how to scale “things” under what a distribution ”might” look like. This is quite a common dilemma in quant trading, but one that becomes integral to research. A 🧵:
I made ChatGPT make a QP Portfolio Optimiser using CCXT and CVXOPT with a turnover constraint.
This is getting ridiculous now.
I mean it's probably like 70% correct and 40% useful, but the fact that it can actually do this much is ridiculous
I was asked a while ago to write a thread about systematic trading on illiquid assets by
@macrocephalpod
. Be it small cap stocks, illiquid shit-coins or a growing new market - there are a few principles worth following. Here goes: 🧵
@implausibleblog
@campbellclaret
I’ve always wondered why political negligence or like, politicians lying, punishable by prison terms etc?
I mean, the £ impact on families by a mis-sold brexit has caused an infinite issue. If someone mis-sold me a financial instrument for £50,000 - I’d have them in jail by now
It’s often reported that quants don’t have stop-losses. We do, but not in the same form as retail traders, because we have different problems. Here’s an incoherent ramble on some points🧵
A few people seem to be missing the point here - Citadel do well in everything they do. It’s not like they just do well in equity l/s or m&a or quant or macro - they do well in everything.
I repeat: it is very difficult to achieve what they achieve year on year
On a long journey home so decided to write a short thread on mean reversion for beginners. It’s the classic statarb strategy that works (almost) everywhere. Here are some messy thoughts:
Citadel is a multi strategy hedge fund. Citadel Securities is a high frequency trading firm. Citadel’s HQ is in Miami, after big Ken got fed up with Chicago.
Follow me for more community updates.
Citadel is a high frequency trading firm in Chicago.
It was founded in 1990 and now has more than $50 billion under management.
Citadel is known for it's ruthless culture.
But it's got really smart quants.
@SkyNews
“Under the agreement, air traffic controllers will also get a pay rise of up to €18,000 (£15,500), 18 additional days off annually, and retirement at 59.”
Meanwhile us brits….
A short thread on cross-sectional normalisation in quant finance. Lots of people use quintiles or deciles, lots of people rank, some people do other stuff, some people do none! A 🧵:
@quantymacro
Disaster. Don’t do this in your 20s. In my 30s now and I’ve seen people who did this get torched/overrun/burned out/depressed. You need at least 3-5 evenings a week with friends: from 7-10pm. Be it sports, dating or whatever. You can’t be creative if you’re behind a book all the
Dad and I go for 5-10k walks often. Last year he started slowing down with angina, and had a heart stent closely after. More than a year after the whole debacle of treatments and therapy, we just walked 10km and he yapped the whole time - no pain, no pause, pure recovery. Big W
Michael Platt's BlueCrest is up 33x since 2016
(by comparison, Citadel's Wellington, the crown jewel of the multi strat hedge fund world, is up 3.8x)
Only one way to get Platt's absurd result
Front running (read cheating)
Anyways, TLDR: ML/AI doesn’t work. Crypto is complicated. Use age old strategies in new ways & new markets. I’ll do my best to answer any questions and apologies if I missed anything out. The above is already quite lengthy!
Ok so this is getting real.
Since NP died, I decided to launch ! Work is in progress, but sign up to keep up to date - and by early Q2/3 we’ll get the full website out.
If you want to join in the fun, DM me!
Sharing appreciated!
Given its recent demise, should I launch a replacement of nuclear phynance? A forum for students/juniors & senior professionals to ask questions in a safe & private manner
My one piece of advice for junior traders: lean into uncertainty and learn to love the punches. It’s the process that makes you a superstar; not the alpha, nor the fund you work at.
Hard work pays money
Market neutral trading is tough for several reasons - but these black holes are usually glossed over when you begin. Here’s a short thread on some problems you face along the way: 🧵
I’ve been asked by
@PtrPomorski
to write a thread about how a quant constructs final $ positions from signals. It’s a nuanced discussion because it depends on what your beliefs are regarding an alpha itself. A 🧵:
In this thread, I've tried to put across the importance of vol when comparing performances.
Within the industry, vol normalisation & vol-targeting is one of many ways to control risk - I don't use it all the time (for many reasons), but it's imperative to know!
Finance personalities are the most diverse: hands down. With diverse backgrounds, but acute aptitude, it’s full of personality. In this thread, I’ll explore some common personality types you might encounter and how to navigate interactions with them. A 🧵
“They say technical analysis is BS but what’s momentum? All price based and technical, no? What’s mean reversion? All price based and technical, no? Not sure if you’ve heard though that momentum is literally everywhere - these guys at AQR wrote about it. It’s mega important bro”
I’ve been asked to write a thread on how to validate systematic strategies. It can be difficult because each strategy is slightly different in nature, but a bit of maths can go a long way. Apologies if I’ve missed anything, and if so, please leave a comment! 🧵
Now that the strategy has similar risk to btc...it makes significantly more than the benchmark! Would you still invest in Bitcoin? For a similar amount of risk, you could make 6x as much!
@Vertox_DF
I would like to see a single alpha; that’s uncorrelated to all my other alpha’s, that’s never featured in academic history.
With high capacity, low transaction costs, high efficacy, and high breadth.
Not asking for much man.
Without an understanding of vol, the risk you take can be hard to determine (note the snaps of vol in bitcoin). So when you compare a seemingly low-vol strategy to the benchmark, if liquidity isn't an issue, you scale up your low-vol strategy to take the risk wanted. Easy!
Bootstrap methods in finance are great to play with assumptions and see how robust your results are to fuckery. It’s a much quicker way to create confidence intervals on estimates, and empirically understand how “significant” an “artefact” may be.
Agh fml - I should probably also say great data resources are Yahoo Finace, the Binance API and also Alpaca. QuantConnect is also good.
SO. MANY. RESOURCES.
I’m going to write a thread on some key personality traits you meet in finance this weekend, and how to manage/deal with them. It’s been one of the hardest parts of my career: the maths, if anything, is less onerous. Let me know if you want any trait covered.
Firstly, make sure your technical skills are sufficient. I’d recommend reading Active Portfolio Management (2), doing loads of Kaggle exercises, understanding distribution theory + bootstrapping methodologies, not to mention being proficient in coding. Pandas is v helpful.
@AgustinLebron3
A PhD requirement is probably the biggest failure of Quant Finance in the past decade. Winton, Aspect and AHL have more PhD’s than risk adjusted returns.
Guy has rented my dad’s apartment: turned it into
@Airbnb
; we try kick him out; going through courts will take 12m & $$$. Meantime, are like IDGAF, so the guy will earn 12m
@Airbnb
rent; not pay anything to us & all refs were forged. Fucked.
The UK is so screwed: a recent article highlights how local councils invests surplus cash. They run internal funds: a mixture of real estate, PE, and other adventurous stuff; all whilst lacking in resources and skilled personnel. It’s ok though, investors don’t need training, as
Now, the fun stuff! ALPHA! Creativity often comes with the problems. I always recommend juniors to read the factor zoo papers to understand how much research already exists () and to think around these topics.
These charts are always shown for the US only. You guys seen European or UK or Asian index levels for the past 10-15 years?
The US juggernaut never ends, but not all beta is equal.
Fun thought experiment: take a strategy that performs identically across asset classes and your annual Sharpe ratio is 18.5% higher in crypto. Why? Because it trades every calendar day and not every business day! More samples, more $
(365/sqrt(365)) / (260/sqrt(260)) = 1.1848.
People are complaining that the returns are < market….investors can lever their investments if they want. Long term spx Sharpe <0.5. Suppose a sharpe of 2 for a fund…at the same vol; the fund makes 4x the market!!!
So as an investor, you want market vol? Sure, here’s a wrapper
Was mentoring a new kid this morning: the kids story was tragic and his goal is honourable: the definition of poor, smart and hungry. All he wants to do is set his mum up.
Seriously - we need more 1:1 mentoring in society. If not for them, your soul will thank you.
Note: this will be quite simple for most quant practitioners. This thread is aimed at juniors/newbies who aren't familiar with leverage, vol normalisation & broader industry mechanics....
Back in the motherland.
Not a unique story but the mountain my father climbed, to get out of the village, to running a successful business in the UK, is absolutely insane. Not sure my future kids will even comprehend.
Big up the fathers out there.
The past 2 years have been an insane time to be a trader. My paranoia levels are through the roof as markets keep falling apart. The top 10 things I learned this year to make $$ 🧵👇
1/n