BlowFish

EMC2 Hope you are still around. With 3 I presume you allow price to confirm which direction it wants to move in and the enter on a pullback? In the case of 4 are you sayin the breakout will tend to be in the posited direction to the extreme? i.e against the anxious participants? Thanks.

Are you implying 'elaborate' markets have less inefficiencies and so are harder to find a statistical edge in? Are elabroacy and liquidity always correlated?

I think you might have misconstrued what I said Firstly it was a question rather than a statement. It was not to discuss the merits of paper trading but to suggest that if you experience emotional responses whilst simming they are unlikely to be simple fear of financial loss. Neither did I comment whether fear was 'healthy' or not. Not being argumentative here just clarifying I too tend to paper trade pretty much as I trade for real (for better or worse). However much you have convinced yourself to 'do it as if its real' if you are experiencing anxiety in sim mode this is probably indicative of other issues. Fear has many many manifestations Only you will be able to tell whether its 'healthy' or not. At some stage it is likely that most will experience the actual pain of financial loss. Tha isn't going to happen simming. The only way to really experience the 'feeling' of a string of loosing trades (and the attendant drawdown) is actually experience it. But that wasn't my point either really. Sim might (or might not) help sorting out a whole bunch of things without the pain (and fear it tends to foster) of losing money.

IB data has been discussed recently elsewhere. What they do is aggregate ticks however in my experience (testing various futures for a couple of weeks against Zenfire) the volume is pretty spot on. Depending on what you want to do it is perfectly adequate. For example if I can use constant volume or time based bars I often use IB & multicharts - if I want to process every single tick or need atick chart I'll use something else. To be honest most of the time it is pretty academic.

I used to a long time ago. I quit as I had data mnagement issues (holes would show up in my tick database). It was with regret I shelved it. The main criticism is that the scripting and programming have a steep learning curve. It's very powerful though. I am sure that the data issue is sorted now. Their tickprecise feature allows you to do stuff that can not be achieved in other packages. I agree broadly with what sevensa has posted about Multicharts. Having said that version 3.0 is reasonably stable and despite niggles it does the job. The beauty is that it compiles ELA's. There is a reasonably broad range of datafeeds. Ninjatrade with Zenfire datafeed/execution engine is un-beatable at the price (free if you don't need dynamic DOM and multi-broker support). Probably every bit as good data and execution wise as TT. Charting and scripting is not bad at all (but with a bit of a learning curve, not as bad as Neoticker). I use the last two currently depending on what I am doing. I would grab a trial MC which would probably have you up and running in no time at all. Ninja is worth taking a look as that won't cost anything. If you are prepared to put the time in with Neoticker I doubt you would be disapointed, it might take a bit of time to get the results you want though. Cheers.

This is a capitalisation issue. Of course a lot of starting traders are under capitalise and/or trade too big. Of course this is likely to produce feelings of anxiety. Knowing the risk of ruin can help. .....Interesting thought, if you experience anxiety/fear when paper trading the issue is unlikely to be fear of financial loss??

You need to put "professional trader" in quotes if applied to Gavin the Huckster

There used to be a company that sold historic DAX data pretty in-expensively but for the life of me I can not remember their name. You could search the omega list archives at purebyte, maybe that would turn up something.

How about:- Hi, I am xxxx. I have been trading for yyy years and am interested in intraday trading the YM using price based S/R levels. I took the Tom Busby seminar which I found valuable but wish I had found this site sooner as I see most of what I learnt there is discussed here in great depth for gratis! Look forward to talking about these techniques.

Many years ago I bought a rather expensive course that had this as one of its premises. That was based on 1 point stops and 10 point targets. A partial was taken at a point/point and a half if memory serves. ES was a little less volaitile at the time. I cant help feeling that using market structure is a better approach. For example when the ES is averaging 30 points a day, swing size is different to when its averaging 10. Taking account of what swing sizes actually are is likely to help. Are you going to sit through 5 6 7 8 and even 9 point retraces to get your 10 points? If you are why limit yourself to 10 points? I always see risk as a dynamic thing (un booked profits is money at risk). Getting stopped after being up 9 must be a bitch. Especially if say 8 points into the trade you notice you are at major S/R. Alternatively (imo) measuring previous swings to get rough targets is likely to yield better results (AB = CD type of stuff). Or simply using next higher time frame S/R for exits. I guess you could 'optimise' (curve fit!) the 10 point target with some sort of back testing. I guess the appeal is simplicity. You might be interested in Eddie Toppel's stuff too. Even simpler, trade a fixed portion of the day (when trends are likely to develop) enter in a random direction with a fixed number of points stop and reverse, its an always in approach. You could probably improve things with a simple directional bias (e.g. enter long if above the open short below). Essentially a trend following system so lots of smaller losses to get the big win. Benefits from adding contracts to winners. I never understood why people do all the 'clever stuff' on entries only, personally I would rather trade a random entry with a good market structure based exit than a market structure based set up with fixed target exit. (I think). Obviously things tend to improve with both.

Yes I was thinking the same thing if a mod wouldn't mind?

Well another day another algorithm this one based on Terriberry's paper 'Computing Higher Order Moments Online". This one is realy neat but I don't fully understand how to weight it properly. I wonder if you would mind posting todays ES Ondrej? This is what I have but I know (well am pretty sure) the 3CM isn't weighted correctly. Its the magenta line. At least its the right shape now (I think). Cheers, Nick.

Recognising an 'edge' is probably what is difficult for many, maybe because they don't really know how to 'test' it. Or even going back one step further are not prepared to document it adequately - making testing impossible. Of course this leads to the behavioural problems of the title, it is impossible to execute consistently with the right mental framework if you have not got confidence in your plan. I think if one starts being a bit more rigorous in setting out what they are going to do an 'edge' is more likely to present itself. Then there is the complexity issue. There are some pretty simple ways to trade but there is a tendency to want to complicate things (to improve the edge maybe?). Result being either confusion of the trader or diluting what is effective. Another issue is not truly accepting that trading is a numbers game and so abandoning perfectly good 'edges' in search of something better. I've done all these things and have to be pretty disciplined not to do them again.

This is not my experience comparing it with Zenfire on various index futures. What they do do however is aggregate ticks however the volume is accurately reported e.g. If you have 10@6550 followed rapidly by 4@6550 they might report it as a single tick of 14@6550. In my experience on the instruments I look at it all gets reported. The situation might be different with say equities (which I dont trade) but I think I would have heard on the grapevine if there where issues. I used to hang out in a room where a lot of comparisons where done between feeds as people changed platforms and stuff. In fact the recommendation would be to trade constant volume bars rather than tick charts with IB. (that is if you need something other than fixed time charts). There can also be issues with backfilling. They essentially 'throttle' requests so it takes a while if you want to load a lot of historic data. The dreaded "pacing violation".

Thanks for your continued dialogue! Yeah its wrong It is a continuous algorithm though so it doesn't use loops at all. It is based on an algorithm by West the weighted incremental algorithm here http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance. I think its real close now but something is a wee bit out. It will run on months of ES tick data though! Its a pretty neat and not too difficult to understand. I think the logic is pretty good as everything is OK except the skew. My hunch is that its the last bit where I divide (n-1) * sumweights. Not sure why though n = 0 foreach x in the data: if n=0 then n = 1 mean = x S = 0 C=0 sumweight = weight else n = n + 1 temp = weight + sumweight S = S + sumweight*weight*(x-mean)^2 / temp // sum of weighted ^2's C = C + sumweight*weight*(x-mean)^3 / temp // sum of weighted ^3's mean = mean + (x-mean)*weight / temp sumweight = temp end if end for Variance = S * n / ((n-1) * sumweight) // if n = 1, omit n/(n-1) SD = Sqrt(Variance) Skew = C * n/((n-1) * sumweight) / cube (SD)

Ok here is the same chart with the sign flipped. (i.e. Vwap - Skew * SD plotted). Thats looking kinda close. Could you post todays ES if you get a moment.

Thanks, I do think the PVP presents some problems also, I'll discuss them later. I am really pleased with the new algorithm as it does not seem prone to numerical instability and overflow errors even when cumulating ^3's. Having said that the skew calculation looks like it needs a tweak compared to yours. I enclose my chart for the same day. I'm really not quite sure what needs doing to be honest. Edit: Didn't Pearson suggest 3*(mean-median)/SD? I wonder if VWAP + 3(mean-median) might be useful?

It would probably would be a lot easier to list those that don't offer tick bars. It would likely be a short list as most packages do.

Well I have made some progress by completely changing my variance algorithm and basing the 3rd central moment stuff on that. Ondrej could you post the odd chart with the whole days action squashed up? I guess really I am going to have to do a long hand algorithm to test against.

And of course there is GJ closing out positions for holiday cash. I would guess planetary alignment is not highly correlated as there is no central exchange for FX and hence not a single location to calculate aspects from. Hehe I am thick skinned and not one of the industrious anyway. There are one or two here that are pretty rigorous and a bit more 'quantitative'.

I did pay around using a game pad for scalping many years ago (A buy B sell). Yes, I was young and foolish. I am pleased to say that I am now old and foolish.

I think you might be judging those that hang out here rather harshly and with a broad brush. It's hardly difficult to test Granger causality with things like Matlab and R having inbuilt functions for such techniques, I'd go as far as saying its trivial. It seems that there several retail traders that post here that have the smarts, gumption and tools to do such work. However I am not sure any trade spot FX (probablly too smart) unlikely they would devote time to prove or disprove a theory that would not have any benefit to thier own trading. Actually seems to me the best one could hope to do would be disprove the theory as there might be a third (unknown) process driving both series and that's one of a couple of places where Granger breaks down (based on my admittedly limited understanding).

Just before everyone agrees to disagree perhaps I might a solicit opinions on a couple of random thoughts (this is just brainstorming, Im certainly not putting anything forward as truth) Is there any value in looking at the futures contract volume? Obviously closely correlated with the underlying spot and centrally traded in a proper auction. (Not sure why retail traders don't just trade the future to be honest). I wonder if some find value in spot 'ticks' for similar reasons that traders watch thinks like the NYSE TICK. This represents the number of stocks ticking up minus the number of stocks ticking down on the NYSE. Some use it to gauge sentiment in different (but correlated) markets (like the S&P). Actually you could probably lean against it trading gold or oil. A fairly abstract beast but some keep there stops tight when extremes occur. I buddy of mine uses astrology to trade (profitably). Not much evidence I can see that there is much correlation there (though he tries to persuade me otherwise), having said that he's pretty well accomplished in other techniques. I think many tools that even experienced and accomplished traders lean against may well be little more than crutches. Having watched Walter trade futures last year (in the TL trading room) he clearly has a great read on price action enough to trade well 'naked' (without indicators) is my guess.

How about the European open in your evenings?

Yes I understand all that but to get a partial fill is so unlikely on the ES you may as well discount it. You wrote "I've had plenty of partial fills " I think that was a mistake? Partials are virtually impossible (on the ES) unless you are trading massive size or submitting orders one lot at a time with a large gap between orders. (so your order is not all in the same place in the queue. ) To get a partial your order not only has to be at the front of the queue at the absolute tick that the market turns - but the size of that last tick has to be smaller than your total size. Not only that but price has to drop from that level and not return. Chances are minuscule in the ES though they would obviously rise if you where trading large size...hence my joke about you trading 1000 at a clip. Of course in thinly traded markets partials are common.