xetra777のブログ

Avatar

The silent or unknown secret of 90% active management profits

Sorry, I write so little and short, but I;m overwhelmed with never ending tasks, forgot all about self-marketing :-D So, the majority of profits comes from fat tails of distributions of returns. Miss best X months and you miss almost everything. Problem is they are generally unpredictable* and as a consequence this is probably the biggest drawback of active management, because you can easily catch those best X months by buying and holding.
* At least to my knowledge. You can probably aggregate a…
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

Why prices move or 2 types of strategies

Prices go up because someone (a group of) agrees to pay higher price.. It is mechanisms behind so called momentum. When "crowds" gather together and want something (whatever reason), and if they believe they would benefit from it (greed component), prices tend to move for extended periods of time. And vice versa.
If you can determine when some particular group would sell or buy (e.g. have a some information - signal), you can benefit from this belief. markets usually have weak signals as they ha…
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

4 types of traders (according to Micahel Harris)

From highest to lowest probability of failure.
1. TA beginners.
Use lots of TA indicators, don't really know what they mean and whether they generate some real signal over noise.
2. Advanced technical analysts.
Experience gave them knowledge on how to deal with discrepancies of technical analysis, they may use less of TA indicators or not use them at all, relying on support &resistance or something like that.
3. Quantitative beginners.
They realize technical analysis isn't correlated to how pric…
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

Projects Quantrade & Ufunc

Updates.
1) I had developed no B.S. free *quantitative* trading signals service at https://quantrade.co.uk.
Tired like hell with countless hours, bugs and trying to fix performance issues, need to move further, so the app will stay on beta until I'll fix remaining few problems.
2) My further project is https://ufunc.com, which will account for what I had learned building Quantrade.co.uk. What is ufunc.com? Most probably it will use machine learning to generate more frequent (unlike Quantrade.co.…
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

4 types framework

After reading several hundred trading books I had come to simple 4-types framework what defines success. Here it is in short.
Negative alpha:
1. Bandit process = leverage * (times) no information advantage.
2. Idiot process = no leverage * no information edge.
Positive alpha:
3. Not successful = no leverage * information.
4. Successful = leverage * information.
Information advantage here can be anything that's most probably is real ("true"). Leverage here can also be anything that empowers usage…
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

QProb Quantitative


After few months of work a week ago I have released world's quantitative traders/ analysts blogs summarizer @ https://qprob.com
Hope soon will enable full text search across them all & other things (should be fixed).
Technologies/ platforms used: Django/ Python, RabbitMQ, Celery, Nginx, uWSGI, memcached, MySQL, Linux, etc.
Above pic is real time sentiment of all posts going into database.
Project in wait list after this one - quant strategies signals (somewhat web based).
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

Rare update

Hey girls, guys. Due to limited time I am able to only update my blog at www.talaikis.com
Best wishes
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す
Avatar

Let's see

I even forgot I have Dukascopy social, haha :-D OK, wanted to try automated championship, but unable to share strategies publicly, so if I'll find time, will do something with manual trading competition. I'm short EURUSD from Friday on another account, but that's short in time also. Nothing else to say as no opportunities are in my watchlist and need to see other pairs.
全て読む
翻訳する: 英語 オリジナルを表示
もしくはコメントを残す