BlowFish

I would say if you have no programming expertise easy language is by far the 'easiest'. It's not really anything to do with the syntax of the language but how you 'interface' with the charting program (to read data series, call functions/access methods, and to plot results). Imho the mechanisms to do this are trickier in meta and ninja (by quite a significant amount). MT4 is fun and colourful but has the drawback that it tends to be supplied by bucket shops rather than brokers. It is a 'locked in' platform I use it (actually I use all three) but for nothing 'serious'. The same could be said about tradestation but multicharts (which compiles easy language) gets round that. There are a few specific little things that only one or the other will do that are unlikely to effect you (unless you are really pushing the envelope). For example only NT has access to the full order book (outside best bid and ask) programatically.

Everyone should drop what they are doing and help you? Right ho I'll spend this week learning the syntax to CQG's scripting language. Maybe you should learn to help yourself. Also maybe you should have posted all the information and steps you had taken at the start. 1) Check the thread that discusses the algorithm in depth. 2) send that algorithm and pertinent comments to CQG 3) Ask them to check they have implemented the algorithm correctly. 4) Profit. That's how a Ferengi would do it anyway. CQG is a premium service, I would be p*ss*d off if I wasn't getting premium support. Thanking people that have tried to help rather than insulting them is probably a good plan too. Oh and all spamming your half arsed pleas for help will likely yield worse results, it 'turned off' Tams who is quite possibly one of the people who could have helped.

I tried despite the multiple posts, but I think I am done. Of course CQG would be the place to go. Perhaps a mod could clean the other threads?

NinjaTrader 7.0.0078 for new beta testers. Friday April 3, 2009 5:00 PM EDT Just a bit of wishful thinking

Actually looks like there is just about enough information to make a vaguely informed decision (I would pass myself). It's not the best site ever but follow all the links and read all the guff and you should get an idea what it's all about.

I am increasingly using 'cloud' storage (out on the ineterweb) for things I consider important. Having said that I am of the school of thought that it is good to reformat and start from scratch now and then. Really what I consider important isn't really....There are a few indicators that took blood sweat and tears to create (even though I don't use them regularly). There are odd web clippings that are really neat (but I never read them). I even have tick data going back years (of instruments I no longer trade). Identifying what is important to you and having a strategy in place to protect it is worthwhile. Backing up everything does no harm either . Murphys law would suggest that as soon as you discard something you will need it!

Many people use the FIX API/message protocol. I think this is probably quite 'high level' as I think other exchanges can be accessed with it. I have it in my head (not sure from where) that it is a PAT systems gateway protocol. Eurex also has a "MISS-based trading interface API". Which is native. The fastest is via Enhanced Transaction and Broadcast Solution. I believe that is only offered via leased line to the exchange.

Welcome! Take a look at the market profile section here, lots of resources mentioned.

Thanks AK, Actually I think it is always 'current'. Obviously adjusting a property portfolio takes a bit more than a mouse click! Any instrument that lets you hedge (even if you are just a simple home owner) is worth investigation. Cheers.

Just an aside - Failure to Traverse was a term coined by Jack Hershey (or so I believe) to describe when price does not make it to the 'other side' of a trend channel.

I could be wrong! but the common mistake is that people dont go back and divide every piece of data by the new total volume. I don't know cqg well enough to tell but there is normally a loop but I don't see that.

Sounds like you are on the right track with looking at money management strategies. You can take something pretty marginal and show a profit with good MM. In fact that was really the heart of the original turtle system, starting small on a break out and pyramiding fairly aggressively. The signal was pretty much an x day breakout (from memory) nothing sophisticated. Something else occurs to me.... essentially you are fading the turtle system right? You might want to look for something called 'Turtle Soup' and I think there is even a 'Turtle Soup +1'. It was a system (by Linda Bradford Raschke I believe) that does just that (fade turtles). I imagine that might be pretty interesting to you. Might be in her street smarts book which is (unusually) a book I don't have!

Not really to be honest I don't know CQG script. If you follow through through the threads there is detailed descriptions of the algorithm in one of them. Did you get the same closing prices (or similar) for the VWAP and bands? Edit: a cursory look would suggest that things might not be correct. Every time you have new volume you need to re calculate the weighting using the new total volume for the whole day. This is discussed at some length in one of the threads as it is a common mistake that people have made in the past. It is probably in the thread where SD is first introduced.

I am not sure either. A screen shot and a precise description of what you are doing would be helpful.

No worries - i notice there is a timer for TS in the indicator section here. what did you go with in the end?

Ok here are 1 tick and 2 min vwaps for the whole days session (Friday). Note the early hour is included. Also note that the time scale is UK time (we don't change to summer time until next weekend so a confusing part of the year). I could have my session times off by an hour I am not trading the DAX right now so haven't checked. Let me know if the session dosen't look right and I'll make adjustments. Things could vary a little with different data sources and on the 2 min how you decide to average the value for your 2 min bar. Importantly you can see both 'work' in so far as price tends to respect the lines. The pvp is a whole other debate, I have some reservations and observations there. That's for another time.

There is a reasonable amount of quality stuff available for free. Sometimes it is a teaser for product, nothing wrong with that of course. Sands does sell courses but if the free stuff is worthwhile who cares? (imho). As an aside I am not sure if you know about the 'drama' between Russell Sands and Curtis Faith. If drama amuses you now and then its worth checking out if not pass on by.

This is only an hour and a half in (or so). Left chart is a 1 tick chart so 100% precise, right hand is 1 minute. You can see there is a wee bit of a discrepancy as you would expect (volume is assigned to the average price of the bar). I'll try and do one at the close but will be travelling later. There is a couple of seconds between the screen grabs as stupid Multicharts wont save the whole workspace. This should let you see if you are close. Cheers.

Actually only 1 guy assumed that and they had a post count of 1. Is posting two days results on a simulator really that much of a "good thing"? Seems pretty random to me but hey what do I know? Anyway if you like that sort of thing you will love this thread http://www.traderslaboratory.com/forums/f103/trader-p-l-2009-a-5521.html where a couple of people post there live results pretty much every day. Btw Jugador hows it going mate? Are you still hitting it pretty consistently?

I dont use CQG but SD bands on weighted data are notoriously processor intensive. Having said that running the calculations every 2 minutes is a snip for most programs (Ensign, Ninja, Tradestation etc.). I wrote my own tradestation indicators that use a non iterative algorithm so no problems with jerkiness even tick by tick on a months worth of data. How many days data are you loading ? If you are only re-calculating every 2 minutes you should have no problems....I would contact CQG as I mentioned before.

I didn't get that impression. Are you still struggling? Did you contact CQG?

Pretty sure thats MetaTrader 4 from MetaQuotes, a popular platform with the FX bucket shops. It's pretty neat for charting but pants for trading.

Good job Frank. Another small programmers tip using the highest and lowest functions will iterate through all the bars each time you run through the code. The construct you used previously comparing the high now with the highest high previously is a more efficient way of doing things (not a big deal in this instance but a good programming habit to avoid loops if possible). Thats the fun thing about coding, many ways of doing things. Not trying to be critical just a 'top tip' to pass on. You could use your date/time logic with your previous high low logic (or not) if you so desired. Interestingly the code will look a little longer but will run tighter. On another topic I wonder why one would exclude Sunday Night (or even Monday morning) from out of hours calculations? They still represent a level where the tide turned.

I'm rather disappointed TL does not make tea yet

Actually Chads post reminds me there is some Easy language code knocking around called wavetimer or bartimer or some such. It prints text next to the bar that counts down to bar end. You can set a 'trigger' where the coulour changes when you get particularly close to new bar time. Might be of interest to you as it should work with OEC. It is pretty rudimentary but if it is of interest I'll dig it out.