MFT & HFT | Views my own. Not financial advice.
the orderbook
- Tweets 15,062
- Following 1,634
- Followers 70,187
- Likes 10,691
Pinned Tweet
Wrote a free article on some lessons I’ve learnt building successful QR operations and how to build successful pipelines
4
119
12,135
Hawk Tuah recently went viral for her rant on the overuse of advanced machine learning models by junior quant researchers.
“Junior researchers prefer non-linear methods and complex black boxes despite the limitations of these models to deal with the high risk of overfitting”
189
3,787
69,499
2,751,444
You shouldn't ever try to brute force "read" a textbook. You read sections, do some experimenting, go back. It's reference material. It's not a regular book. Trying to read the whole thing is a waste of time and you'll forget it all.
168
466
8,130
297,853
‼️ Quant Roadmap 2024/25 Ultimate Edition 🚨
Containing:
- Hundreds of resources
- GBs of free data
- Guidance on strategies
- Much more!
[link below!]
66
268
1,843
316,493
Probably the closest textbook to the knowledge you would learn on the job. Really in its own caliber. If anyone knows similar books please share below
27
226
1,654
295,099
A full thread of all my textbook recommendations.
There will be 3 sections. Research, Options, & Programming.
...
32
284
1,458
317,856
As many of you may know I publish a PDF every so and so often that is a mix between a massive link dump for learning everything quant and generally a bit of advice and guidance on how to learn it. For pros and beginners alike.
cdn.discordapp.com/attachmen…
40
252
1,387
Exile was basically a death sentence back then. You forfeited all property, land rights, etc. You were kicked from your guild so basically couldn’t practice your profession in any legitimate settlement. Most communities would refuse to let you in, and no one would touch you.
It was basically saying go to the forest and live there or become a vagrant. A lot of exiled people back then would just kill themselves and get it over with.
It’s not quite the same thing imo. It was basically a humane death penalty for those times, not just “get out of France you now have to go live in Spain” as we’d imagine it today.
6
22
1,190
45,362
Andrew Tate in the orderbook. Buying 13% of supply 10 mins before making a video pumping a memecoin.
etherscan.io/address/0x6131b…
Trying to hide the routing, wallet was an owner within an hour of token release.
0.007 -> 0.16, sold an hour later for 200k profit.
59
95
980
400,772
Absolute gem !! - a lot of practice-focused sections which you rarely get for HFT, normally all integrals
10
123
875
132,414
Caved and purchased the hardcover after enough flicking through the pdf @__paleologo :)
14
61
727
34,660
This is a pretty important primer. Links below (also background knowledge textbook provided)
19
79
738
126,744
|￣￣￣￣￣￣￣￣￣￣￣|
| Your “alpha” is just |
| levered beta. |
|＿＿＿＿＿＿＿＿＿＿＿|
\ (•◡•) /
\ /
——
| |
_| |_
23
58
628
78,348
If you want some info on how to do market making in these markets, I have some great articles on the topic which go into detail.
Stablecoin markets are piss easy to MM, if you want some easy market making experience. Go MM stablecoin to stablecoin markets. Super non-toxic, can literally do it manually. I made money doing it manually at one point.
8
36
598
105,202
Imagine you are the CIA. Your job is to not get caught. You pitch this idea to your boss. What does he say?
He says are you fucking stupid.
Why on earth would anyone other than an idiot normie intentionally add clues it was them.
23
11
525
73,799
It's often joked that you start off doing arbitrage and end up as a market maker.
Here's my take on why.
A thread 🧵
[1/14]
11
60
539
180,993
A really great primer textbook, but with the bonus of covering areas I rarely see published.
15
72
506
56,742
The clip of Caroline Elison saying “we don’t use stop losses”, “we don’t use technical analysis”, and “we use a lot of elementary math” is to the untrained eye terrible, but most practitioners would agree with all 3 of those
Digging through each:
…
19
47
490
Finished writing a full guide on pairs trading, specifically finding great portfolios. It’s a pretty comprehensive dive into the topic.
Pairs trading is an area I worked on for a while, but long enough ago that I’m happy to share my insights on it
quantarb.substack.com/p/pair…
15
75
505
63,478
GoDaddy and Porkbun just frontrunned me for a domain name, seconds after I searched for it.
I eat bots like this for breakfast, every damn day in the order book. You are nothing.
The solution is clear. Code a domain search spoofing bot. Two can play at this game.
30
21
485
82,311
Good reading material
roycheng.cn/files/riskModels…
alacra.com/alacra/help/barra…
top1000funds.com/wp-content/…
alacra.com/alacra/help/barra…
twosigma.com/wp-content/uplo…
24
40
482
50,178
Reminds me of Positional Options Trading, it's a textbook I could read all day. It's all the cool nuances and ways of thinking about things that makes quant interesting. Like I will sit and read a textbook made solely out of formulas - I just won't enjoy it much
11
57
451
74,575
The 10%+ average return for stocks is a lie. It uses the massive bull run in the past few decades to calculate it. Going 100 years back many of the top economies stock markets would have done terribly, using US returns is survivorship bias. Speaking of which,
…
29
28
451
164,429
Algovibes is really an amazing YT channel for getting started with trading algorithms.
I've seen many and can easily say it is the best.
(part-time larry is great too!)
17
36
439
47,283
1 textbook for theory, 1 for practical, this is practical. Regressions should be a key study area for you.
8
53
434
59,508
In case anyone doesn't know... if you want to scrape Twitter, or any site that doesn't let you scrape it, but has an expensive API:
Use RapidAPI. Anyone can create their own API and sell it.
I can get all the public functions of the Twitter API for $50/mo and 100k requests p.m.
10
19
418
54,413
Bollinger bands are to this day an amazing quant tool. It’s just rolling 2 standard deviations after all. Almost any reversal to fair value can easily be traded with bollinger bands as a dirty solution.
14
18
413
58,439
I don’t see why everyone doesn’t finance every Apple purchase. 0% interest for 24 months. That’s like 5% off. Free money.
18
7
416
39,307
Just start building trading algorithms.
They’ll suck at first, probably for the first few at least but eventually you’ll get the hang of it.
Don’t need to be a good coder, you’ll figure it out by doing it.
Iterate, and improve and things work
12
29
414
40,362
By a mile the most detailed information you can find online without working at a shop.
Coming from someone with real experience running HFT operations & has had pnl attribution under their name on a top HFT desk by volume
If you want some info on how to do market making in these markets, I have some great articles on the topic which go into detail.
9
11
425
52,852
1. Trading is extremely hard which is why I don't trade.
So instead I sell a course on how to write a basic backtest and faked the rest of the pnl using a paper account
...
10
10
411
13,698
In these two examples, both have the same max(pnl) of 10, but one is pure noise, if you optimize on fig 2 you overfit.
If you step left / right you get a completely different pnl in fig 2, but get similar pnl (give or take 15%) in fig 1.
If you can't change your params reasonably liberally and still perform well then your strategy is likely overfit.
If your strategy has signal it will shine through different choices of parameters.
Z-axis is PnL.
This may seem very trivial from looking at these plots but I've seen people make backtests and then you modifying the value of an SMA lookback by 20% and it falls apart.
The robustness of your signal to param changes is a signal for how likely you are to be overfitting.
12
23
396
32,312
You buy physical textbooks because you don't like reading PDFs
I buy physical textbooks because I've read & pirated the PDF and now want the bookshelf flex
We are not the same
19
19
386
24,110
Raising capital. A long ass thread.
🧵👀🕵️♂️🫰🤑💸
Sources, structures, and general tips.
...
11
60
389
92,228
You can just do things in crypto.
1) Min trade size is tiny
2) An AWS (usually Tokyo) server is all you need to be near the exchange
3) non top tier exchanges will usually give anyone who asks nicely a near top fee tier trial
4) plenty of small edges exist
5) exchange data can be scraped for free directly from the exchange
25
12
399
64,285
My simulations show Kamala winning across all scenarios Nate
It's published.
We ran 80,000 simulations tonight.
Harris won in 40,012.
natesilver.net/p/nate-silver…
30
13
379
58,949
Current (re-) reading list. Read 2nd ed. for #3 a while back, reading the 3rd now. #7 & #4 are just huge textbooks - I don't know if I'm reading or re-reading I just pick a chapter. Some I've never read, some I've read 3x
10
55
381
38,623
Here’s a thread of all the material you need (that I’ve written) to learn how to do arbitrage strategies like the pros:
…
9
25
389
49,127
Please anyone thinking of starting a hedge fund read this ‼️:
Use managed accounts. That’s it.
Legal costs for a hedge fund can get close to $100,00, each managed account is only $1,000-$1,500 in BVI, up to $5,000 if you add expensive lawyers (not 100% necessary) …
25
52
367
