Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Well, no. I'm saying that the strategies I'm using for intraday trading no longer have a positive expectancy, so my bet is $0 on those strategies. I'm 75% in the longer time-frame stuff right now, while I work on finding some intraday strategies that work again. So its a question of my ability at this point in time rather than a question of whether short-term trading works at this point in time.
Mike,
Its OK to argue with me or to ask an explanation, but to state some stupid remarks is also OK (we need stupid traders to make money), only I'm out of this thread.
If you look at the strategies I've posted over the past few months you will see there is a method behind my madness :
Faber(2009): Momentum, Monthly, Basket of Futures or ETFs, Long only.
Kleinman(2005): Momentum, Daily, Basket of Futures, Long and Short.
PimpMyStrat: Momentum, Intraday, Individual Futures, Long or Short.
NinjaSpreader: Mean reversion/Market Microstrucure, Intraday, Futures, Long and Short.
NinjaSpreaderEOD: Mean reversion, Daily, Individual Futures or ETFs, Long and Short.
So when they (or something simlar) actually works, that's a basket of strategies diversified over time frame, style, and asset type. One of the next steps would be to find a way to dynamically watch my deltas (or beta-weighted deltas) for hedging purposes if needed.
That is actually a seperate project where I want to get several Ninja Trader processes running on seperate machines to report positions to a central database on the same LAN. If everything decided to be long the ES at the same time, for example, I would definately want to know that.
First idea that came to my head was to have each strategy send its orders to a gateway, then the gateway knows everything. You could even have the strategy ask permission from the gateway before trading, or if you are super wild then the strategies themselves don't contain any EnterXX() commands and instead use custom commands to send orders to the gateway server, which then submits all orders and controls all positions for all strategies.
I will take a closer look at your other strats. I know you've done some great work, and I appreciate it!
On the broker side I can set that up on a per account basis as the broker has a risk package that can limit positions. If I ask for it I can have multiple accounts and a "dashboard" to monitor those accounts (this is using TT or PATS infrastructure) which may or may not come at a cost.
My problem is that I'll be using more than one broker, on a WAN.
That is something to keep in mind, BTW. Your orders always go through your clearing houses' risk system before they go to the exchange. If I have a server in Chicago and my broker is in Singapore... ouch.
So probably best to do a co-lo or get a dedicated box or VPS near your broker/exchange, and run everything there, and just have it send you the info to your home network or wan after the fact, when milliseconds don't count.
HI, I find this topic very important. I made 4 strategy that have a little bit of difference between them and trade on the same symbol all for but not at the same time frame. I must say that the 4 are not corrolated to each other and only trade maximum 1 contrat a day on each strategy. ANd I must say that it work so far. I watch those 4 strategy back in june 2009 and no optimization at all and the result seam to be constant so far. The 4 strategy are pratictly 0 correlation. I think it's the a good way to go. I backtest on minute bar 10 year of data.
Exactly. So the multiple-Ninja-Instances on a LAN (the WAN connections are VPN so it is actually on the LAN subnet) reporting to a database with a dashboard is what I need to do.
BM, I am very impressed with your forums and the knowledge of it's partcipants.
Currently, I am managing 6 Uncorrelated* Forex strategies on NT 6.5 24/5. Trades about 10 rt per 24hrs. 7 pairs using multiple timeframes. A few things from my interpretation.
1. Correlation- currently GBPUSD & USDCAD -94 based on 60min timeframe.
-7 on a 15min timeframe. Overall a moving target and beyond my knowledge to
predict how long this will last. My answer= reduce risk, trade them both with
a few strats on different time frames. Will my bet size be optimal..no but I can
control risk say up to 6 independent positions all risking 0.5% of equity.
2. Position sizing - Since I have no clue which strategy will work today they all get
a feeding of between 0.25-0.5% of total equity risk. Summary = reduce risk.
3. My Landscaper - Discovered an underground spring on my property.. My response
"Damn it". His response don't fight it, work with it.
: