what else in Amibroker shall I pay attention if I want to backtest futures?

[droskill]

Quote from signalstop:


Hey, I'm going by what several people told me about Ami, never tried it since when I checked it out it didn't have any builtin mechanism for fully automated trading. I hear there are improvements in that area.

Don't really know about it for automated trading as it's not my area - I just do long term and swing trading.

The platform I mention and prefer not to specify although it's already mentioned in this thread goes a lot further in helping match historical and real time.

Ok - I'll guess it is TickZoom? I'm not clear why you're reluctant to say the name.

For example, have you figured out that you can't use money management stops like Profit Target and Stop Loss on Ami and most other popular platforms? Most successful traders leave them out of historical tests because the platforms handle them by "mixing" them with the natural trades. Instead those traders select their money management on a spreadsheet by analyzing the trade history. Then they trade the strategy manually and apply their chosen stop losses.

The platform mentioned does that same "manual overlay" automatically by keeping the money management separate.

If you're unclear of this problem it's a question of whether to mix the money management exits with the "natural trades" or overlay them. Mixing is terrible and how all the popular platforms do it. So you get weird results when optimizing and more likely to "curve fit".

Well, I tend to trade without stops for the most part as I've found they generally hurt systems. Profit targets are an area I haven't explored in any real depth. But I'm really not clear what you mean by "mixing" them with normal trades. There is nothing in AB that would prevent you from running a system without stops and then adding in Excel, but I"m really not clear what you think the benefit is.

Are you concerned about being able to, say, be taken out of a trade at a specific price, on a stop, in, say, a fast moving market where you never would have gotten that price? If that is the worry, then yes, I agree that can be a problem. I have one system that uses stops - and I've never had a problem getting out of a trade within a normal slippage amount - and that includes trading through 2007 and 2008 when volatility was very high. But it could happen without a doubt. Avoiding the NYSE specialists certainly helps!

 

[droskill]

Quote from signalstop:

Droskill, another benefit of you using only market orders is avoiding hassels with partial fills. The other platform I mentiond elegantly handles partial fills with limits and stops but you avoid that entirely with only market orders.

Do you do any advanced techniques like variable position size or portfolio trading? Do you do any portfolio trading with either multiple strategies per symbol or a strategy on multiple symbols? If so do you do on-the-fly automated asset allocation position sizing?

That's well nigh impossible on most platforms in real time. Just curious if you found another clever work around for those areas also.

I do variable position sizing in AB - some standard ones like volatility position sizing and some others that I've come up with myself. Not difficult and accurate as long as I'm calculating on the prior close. If I'm trying to size based on, say, the current close, then I have to do an external estimation of what my position size will be based on different outcomes. I can obviously backtest it perfectly, but for the trade, it requires some estimation.

I do a lot of "strategy on multiple symbols" - and I do a bit of "multiple strategies on a single symbol" - I'm working more on the latter than the former. Agreed that most platforms can't do it - what is required is the ability to have multiple systems and then a weighting applied to each system (based on anything you want).

But for single-symbol testing with multiple strategies, I've found nothing better than the flexibility of Excel. Excel breaks down for me, as a tool, when I try and use it for a lot of symbols - it can obviously be done (I know lots of people who do), but for me it becomes too cumbersome.

 

[signalstop]

Quote from droskill:

Don't really know about it for automated trading as it's not my area - I just do long term and swing trading.

Well Ami and others are fine for testing ideas if you have the discipline and patience to trade them manually. I have the discipline but lack the patience.

Quote from droskill:
Ok - I'll guess it is TickZoom? I'm not clear why you're reluctant to say the name.

Well, I tend to trade without stops for the most part as I've found they generally hurt systems. Profit targets are an area I haven't explored in any real depth. But I'm really not clear what you mean by "mixing" them with normal trades. There is nothing in AB that would prevent you from running a system without stops and then adding in Excel, but I"m really not clear what you think the benefit is.


Are you concerned about being able to, say, be taken out of a trade at a specific price, on a stop, in, say, a fast moving market where you never would have gotten that price? If that is the worry, then yes, I agree that can be a problem. I have one system that uses stops - and I've never had a problem getting out of a trade within a normal slippage amount - and that includes trading through 2007 and 2008 when volatility was very high. But it could happen without a doubt. Avoiding the NYSE specialists certainly helps!

Nothing like any of that.

Let me walk through a simple example as to the real reason stops hurt a system as you mentioned you have observed.

Let's say you create a simple system with 140 total trades that makes money but you see based on MAE that a stop loss would reduce risk, draw downs and increase profits.

So you add a simple stop loss to the strategy. Then you are shocked to see that the results are not what you might expect. The strategy might be showing worse instead of better statistics and might even be showing a loss.

But how can that be since you only cut of some losing trades??

Well, if you look closely you will see that you now have MORE trades. That's right instead of 140 you maybe 200 or more trades.

That, my friend, is the essence of the problem. That's because the stop loss exits were "mixed" with your natural exits.

So those several large losses that perhaps took several days before exciting naturally, they now exit quickly which offers the system the opportunity to trade more during the new flat period. Those are most often new LOSING trades to boot.

This situation with all the software platforms forces traders to trade their system manually to apply the money management stops themselves.

And none of the platforms you have heard of except one handles this without "mixing" the exits. Instead this does an overlay.

That can make it far simpler to fully automate a system that previously required a trader to manually handle the money management exits.

 

[signalstop]

Quote from droskill:

I do variable position sizing in AB - some standard ones like volatility position sizing and some others that I've come up with myself. Not difficult and accurate as long as I'm calculating on the prior close. If I'm trying to size based on, say, the current close, then I have to do an external estimation of what my position size will be based on different outcomes. I can obviously backtest it perfectly, but for the trade, it requires some estimation.

fascinating. I will have to ponder this.

Quote from droskill:
I do a lot of "strategy on multiple symbols" - and I do a bit of "multiple strategies on a single symbol" - I'm working more on the latter than the former. Agreed that most platforms can't do it - what is required is the ability to have multiple systems and then a weighting applied to each system (based on anything you want).

Quote from droskill:
But for single-symbol testing with multiple strategies, I've found nothing better than the flexibility of Excel. Excel breaks down for me, as a tool, when I try and use it for a lot of symbols - it can obviously be done (I know lots of people who do), but for me it becomes too cumbersome. [/B]

Yes. Excel or even OpenOffice Calc (<-- which is free) are both awesome for figuring out some advanced stuff.

It's just nice to have a platform that performs all that multi-strategy and multi-symbol work builtin so more energy can go into other things.

But since you're trading manually over longer periods and you have the patience, then you don't need advanced software platforms. And that just keeps more money in your pocket.

 

[droskill]

Ok - still mystified as to why you don't want to name the platform - which is too bad because I think you're right about the example you've given. I'm going to see if AB has any solution, if not, I'm going to have to go back and retest some strategies using Excel to do the stops!

 

[Mike805]

Quote from signalstop:


Let's say you create a simple system with 140 total trades that makes money but you see based on MAE that a stop loss would reduce risk, draw downs and increase profits.

So you add a simple stop loss to the strategy. Then you are shocked to see that the results are not what you might expect. The strategy might be showing worse instead of better statistics and might even be showing a loss.

But how can that be since you only cut of some losing trades??

Well, if you look closely you will see that you now have MORE trades. That's right instead of 140 you maybe 200 or more trades.

That, my friend, is the essence of the problem. That's because the stop loss exits were "mixed" with your natural exits.

So those several large losses that perhaps took several days before exciting naturally, they now exit quickly which offers the system the opportunity to trade more during the new flat period. Those are most often new LOSING trades to boot.

This situation with all the software platforms forces traders to trade their system manually to apply the money management stops themselves.

And none of the platforms you have heard of except one handles this without "mixing" the exits. Instead this does an overlay.

That can make it far simpler to fully automate a system that previously required a trader to manually handle the money management exits.

Are you serious? You really see this "mixing" issue as a problem with the software? You do realize this can be easily handled by writing proper code in every single tool mentioned thus far, right?

What you're talking about is coding a simple lock out mechanism. The problem is behind the keyboard, not with the tool.

 

[FutsTrader111]

Quote from adadadog:

When software shops build software, they have alpha version first for internal testing; then beta version for users testing. The reason using beta version to let user test is because there is liability issue. Software producers are liable for their Software causing damage to users. IBConnector is built by IB to work with amibroker. If any software used for trading and causing loss to traders because of faulty software, traders can sue. Apparently IB does not have enough confidence in Amibroker to have their very simple program IBconnector to be a formal release after 5+ years.

Anything else need to be said?

No shit sherlock. Any software engineer or technical professional that builds software knows this. You make it sound like your the only one on this board who knows what you are talking about. Why state the obvious.

 

[a5519]

This tikczoom spammer is again on board to explain to everyone what trading is. He has been already deleted but now managed to create a new user.
This behaviour is absolutely reliable certificate of the quality of his sw.

Buy, everyone buy, and you will get unforgettable features.

 

[psytrade]

tickzoom delayed my lunch by 30 minutes one day... He knows how to talk.

 

[signalstop]

Quote from Mike805:

Are you serious? You really see this "mixing" issue as a problem with the software? You do realize this can be easily handled by writing proper code in every single tool mentioned thus far, right?

What you're talking about is coding a simple lock out mechanism. The problem is behind the keyboard, not with the tool.

You're right a "lock out" mechanism is what it takes. It just would be nice to use a platform built by traders who "get it" and have all that stuff built in.

The only problem with the tools is that many people use the builtin abilities without "lock out" mechanisms and unwittingly waste hundreds or thousands of hours on strategy research.

Of course, the list of "gotchas" in builtin features goes on longer and therefore takes enormous time for each trader to figure them out.

The tools are awesome, of course, far better than doing it all entirely manually. But the tools are like a chain saw, great for cutting--just the inexperienced may hurt themselves.