Discussion in 'Automated Trading' started by AlgTrd2018, Jul 1, 2018.
0 slippage in ES, huh?
Famous last words.
The way the PnL fluctuates over such a short period of time tells me tens of thousands of trades must have been taken, in addition, no trade quantity count is shown in the screenshot.
Great strategy if you pay no commissions, cqg tariffs, or nfa fees
If his strategy really works, he can go to a hft firm. They pay almost 0 fees.
LOL. Let's see - I program everything in python, don't really know any APIs and might have read the CFE PITCH manual by accident. Yet somehow I manage to be employed and am making money.
Come to think of it, most senior traders/PMs that I know in the systematic community are not hard-core programmers, with HFT guys being notable exceptions. IMHO, the right question to ask is "why do you think you can find alphas that other people have missed?"
why don't u quote my full sentences? Now I copy it again, please read it carefully
slippage depends on the financial instrument and the strategy, ES has enough liquidity and for this strategy, slippage is considered as 0, which is almost the same as real time trading
that is based on facts or results
please tell us the value of slippage you would like to use for ES if u disagree
At present, I've developed my own trading system for automated trading, which is programmed in C# via XX API, and in the mean time, I also use XX as my current main Data Feed. (For some reason, I use the XX to replace the company's name) But I've worked with multiple other trading APIs and Data Feeds before. (I use my own trading system for automated trading, but there are a lot of ready-to-use tools or platforms for backtest) I have got a master degree in computer science and practical programming experience with C#, C++, Python, JAVA, VBA etc. I've worked in a PE and have practical experience and skills developing trading strategies for stocks, futures, options etc.
It has totally nothing to do with "curve fit". That is a waste of time and meaningless.
For some strategies, for example, a basket of stocks, it might be necessary to back test data of many years. Sorry for this strategy I posted, it makes no sense to test that long. No overnight position, the results can be seen within one day......
Maybe this is your own case or experience.
It is not easy to find two identical leaves in the wild world.
Mine is totally different from yours.
For slippage, just try executing the algo manually and you will get a sense of slippage. For ES if you are not colocated when you get filled, expect the fair value of the futures (market maker quote) to have moved at least 0.5 tick away from you because of adverse selection and last position in queue.
Separate names with a comma.