There will be 3 sections. Research, Options, & Programming.
...
1.
Machine Learning for Algorithmic Trading. Great read, but try not to go too deep into the machine learning path. The way they think about features as alphas in chapter 4 is a really good framework.
...
2.
This continues the mental framework used in ML4T and is a really good read. Worth pouring over.
…
3.
This is a great one and I like the pipeline which is data -> alpha -> forecast (EV estimate) -> portfolio optimisation (trades) -> execution.
Portfolio optimisation chapter is brilliant.
…
4.
A bit of help on the execution side, but generally avoid mathematical models and focus on papers that are empirically focused. Meaningful results as well, not just “we observe autocorrelation with X p value” tf does that help me lol.
…
5.
This is basically the bible of statistics. Read it 3x.
If you are having trouble understanding it, there is a beginners version which includes code in R (generally more code / intuition vs. math / rigor)
This is really good and will help your understanding of modern strategies. Key building blocks for designing strategies in the future.
…
7.
Another great one on the topic of trading strategies and getting ideas. I think all of the backtests shown are overfit to make them look better, but it is a good idea to get familiar with approaches / ideas to improve your own creative process.
Financial time series analysis. Some good methods, but I wouldn't get too attached to these ideas as a gold standard. Many of them are a bit too simple or assume too much. Robust methods do provide valuable info tho.
...
9.
Maybe I just really like Bayesian approaches, but I think everyone should be familiar. James V Stone has a brilliant introduction textbook (in addition to this) for bayes which is designed for basically 14 year olds - can't get confused.
...
10.
Stealing this one from @Vertox_DF but it's really good to get exposure from other areas of statistical modelling.
...
11.
This one has a lot of hype, but it also has a lot of generally good advice for backtesting / data cleaning. Avoid the bs quantum computing, but all round should be on everyone's bookshelf.
..
12.
This is another one like the DSP textbook that is a bit more optional. I find a lot of value in these methods / understanding how other fields approach these problems. This one is certainly very interesting and one of my favorites.
...
13.
Having a deep intuition of what your linear regression is telling you is important. Read this and also apply it with kaggle / a wide variety of datasets.
…
14.
This one gives a lot of intuition / deep understanding of linear regressions.
...
15.
Last textbook in the research section. Convex optimization should be in everyone's toolbox. If you want to take it further I've added a second bonus one (Handbook on semidefinite...) which goes a bit deeper.
...
16.
For the options section I'm just going to list them.
Hull - Options Futures & Other Derivatives
All 3 of Euan Sinclairs Textbooks esp. positional option trading
Trading Volatility - colin bennett
natenberg option volatility and pricing
...
The Volatility Surface: A Practitioner's Guide - Gatherall
Trading Options Greeks: How Time, Volatility, and Other Pricing Factors Drive Profits
...
Honestly, I should just concede defeat for options resources because we all know @KrisAbdelmessih has the best resource list:
I'm sure I could go on for a while with programming textbooks, but I'll leave it there. You won't get good at any of these just off textbooks. Implement it, talk about it, write notes, come up with your own ideas, and watch videos / podcasts / other media sources.
..
I recommend checking out @Vertox_DF's pinned tweet, where he gives some videos / where I've stolen a couple textbooks from :)
I'd also like to note that there is a TON of valuable info in podcasts.
Flirting with Models - Corey Hoffstein I've listened to every episode
...
some of the episodes I've listened to 6/7x (because I was implementing one of them, but they're really good).
A thread on all the components of latency, optimizations, & assumptions with modelling it.
...
This will primarily be for HFT, and focus on digital assets, but I will explain which parts are digital assets specific and which parts are not as much of it is generally applicable.
...
So what are the 3 "components" to our latency:
1. Our compute 2. The network 3. The matching engine
It’s the gains in performance you accumulate over time from tuning your strategy and improving it.
…
When it comes to non-HFT, accumulated improvement often leads to overfitting.
Continuing to tune the model once created often leads to decreases in performance other than simple re-fitting of the model on new data that has come out.
Let’s say a new trade has occurred on an exchange, if we have a latency edge we want to be one of the people incorporating it into price instead of one of the people reacting to price changes.
…
As we can see based on this below Pepe, a trade will cause an initial spike before a much slower levelling off.
Where it levels off to (relative to starting point) is going to be important to know as well as both the spike up and return points.
Fill probability analysis is primarily useful when optimizing maker/taker trades.
These are trades where the first leg we make into and then the rest of the legs are takers.
This is a limit order and then market orders the rest of the way (either limit IOC or market)
...
An example of this is triangular arbitrage, where we make into the first leg and then use taker orders to exit.
How can we estimate the probability of getting filled at any given level, and thus use this information to determine the optimal amount of spread to quote?