As with most things, I believe its more about learning to think in the right way than a specific knowledge base. In research in general it rarely the quantity of somebodies knowledge that enables them to make big breakthroughs, its more to do with how they approach problems. Einstein didnt come up with relativity because he knew the most about physics, he did it because he was able to deconstuct the problem in an insightful way.

if you are really new, id recommend http://www.amazon.co.uk/Stock-Market-Edges-quantitative-developing-ebook/dp/B00DY73M24/ref=pd_sim_kinc_2?ie=UTF8&refRID=1K8J7CGJMJ866JVBWMC8
the strategies are pretty basic, and id wager most have no edge, or such a small one not to justify the time implementing them. But its a couple of pounds, and has quite a good methodology that one can learn a bit from. even though they are simple, taking the time to implement the strategies will probably get you thinking in a more research orientated way, not to mention teach you about model building.

99% of what i learned from my research in engineering came from sitting in front of matlab just trying things out. grab a good book on portfolio theory, implement a mainstream model and see what happens. If it works, try to interogate what are the good and bad points, and try to improve it. If it doesnt, try to understand why not. I struggled at first when i started learning finance, because you want a quant 101 book that lays it all out. The reality is these books dont really exist, as 99% of those who develop strategies that really work well can make much more money trading on them or selling them to prop houses than by publishing them and having the edge dissappear. if you can learn to be guided by your own investigations (guided in bits by others work) you will be in a far better place than most.

It normally takes phd students 3-4 years to develop new research, and they do it full time, so the wisest thing to do is to treat the first bit of time as a learning exercise. problem solving without any real guide is a hard thing to do, but its essental to learn, as if you ever to develop a new strategy, there wont be any specific literature to help you. after doing that for a good period of time, you can develop a sort of research sense, which is invaluable too. Most Dr's and above can look at a thesis on new work they havent researched and in about 10 mins choose 3 or 4 lines of enquiry to pursue, not to mention highlight any weaknesses. They cant do this because they have read more textbooks than you or I (I dont think my supervisor had read one in about 20 years), but because they have so much experience of tackling novel and new problems.

Hi Paul,

Welcome to Quantopian!

As a starting point, I would recommend Ernie Chan's book. Ernie does a great job of introducing important concepts, explaining industry terms, and highlighting technical methods. As a reference, here's his EWA/EWC pair trade implemented on Quantopian.

For more information on our API and to get started using our platform, see our Help Docs. Our IDE is Python based - if you are a beginner with Python there are great free online courses from places like Code Academy that can help you get started.

And of course, keep asking questions! Quantopian is a friendly place and members will help with both technical and financial questions.

Take care,
Alisa

Chris - Thank you so much for your insight. I am very new and will be reading the books you recommended, I see this as a learning experience and plan on dedicating many years to this cause. I believe that if I can become a successful algorithmic trader, I can benefit from it for the rest of my life. I appreciate your insight on research and will use it in other pursuits as well.

Alisa - Thank you for all the recommendations, I have added Ernie Chan's Book to the list as well and I am actually working through Code Academy's Python lessons as we speak.

I will try to document the process of learning so it can be useful for the community later on! Any more insight is highly appreciated.

Paul

Hey Paul,

Just something I learned over the years. Usually there are no one strategy that works all the time. It's all up to where in the market cycle we are in or the conditions suitable to different strategies. So I believe that it's very important to understand the type of market you are trading. If you know that you know which strategies and how to structure your trades.

Here is my recommended book list to get you started:
* Algorithmic Trading: Winning Strategies and Their Rationale by Ernie Chan. This is his second book on the topic and much improved over the first.
* Algorithmic Trading and DMA by Barry Johnson. This will give you an understanding of how exchanges work and market micro-structure.

There is also an extensive Quantitative Finance Reading List over at the Quantstrat blog.

Dulguun - thanks a lot for the help, that makes a lot of sense! Algorithms have a time limit basically.

Aidan - Thanks for the books! Will add them to the list.

Also for the newbies like me reading the thread, these are some free courses I plan on taking:

Mathematical Methods for Quantitative Finance on Coursera

Analytics of Finance on MIT Open Courseware

Investments on MIT Open Coursware

Hey Paul,

Welcome to Quantopian! One of the best books on statistics and data analysis, which can be invaluable in algorithmic investing, is The Elements of Statistical Learning by a group of professors at Stanford. The book is available in pdf form for free online. Another great resource is Meucci's Risk and Asset Allocation, which provides a rich and rigorous foundation for portfolio construction and risk management. The Numpy, Scipy, scitkit-learn libraries in Python are becoming more prevalent in data analysis. A great way to improve Python skills is to make an open source contribution.

Best,
Ryan

I just discover Quantopian by random chance and I am so glad! I was searching why performance returns for a particular holding differ from Bloomberg versus Morningstar and I found the Quantopian community! I too am new like Paul and want to learn as much as possible and have already made notes from just reading all the wonderful feedback that everyone gave him. Thanks Quant community you guys rock!

Quant Newbie,
Ari : )