The two CCI indicators have different purposes. I use the 30 min. CCI to indicate overbought/oversold conditions. CCI above average is going into overbought territory while below the average would indicate oversold territory. The 5 min. CCI is used for timing. When it's rising and crosses the average it's saying the downward momentum has eased and the market is moving higher. I suppose any momentum indicator would work just as well.
I missed the first trade because my internet connection wasn't working. I had 3 other trades. Buy at 1309.25 stopped out at 1307.25 -2.0 Buy at 1306.75 stopped out at 1305.25 -1.50 Buy at 1303.75 exited on close at 1307.25 +3.50 Pretty lucky at the end. I'll refresh the data later and include the missed first trade in the summary. Still, for it being a trend down day for most of the day, I didn't get hurt...and that's a good thing. Hopefully we'll see some better trading action within the next few days.
something strange seems to be happening with the CCI average indicator. look at the CCI-30min plot in the first post. on the last bar of the 30min plot, CCI = -89.12 CCIAvg = -20.65 now just eyeball the CCI number for the previous 10 bars. the average should be much lower than -20.65. i have checked other dates with actual plot numbers from tradestation and i see the same problem.
Has anyone coded this strategy up in TradeStation etc to see if it exhibits positive expectancy? It would be a shame if it was a net loser over the last couple of years or more.
Excuse me: I just want to get the record straight. I did trade a variant of your system through my fund (I am retired now). The fund still trades the system and does quite well with it. I do not use it currently for my individual account. My comment is on Page 7 of your previous thread. Steve
As far as I can see is this a temporary success, if at all... This system is losing not only in trends, but in choppy and spiky and flat environment, the Chart on the examplechart shows a market with very nice swings, nicely rounded, excellent. You can trade such a market with mostly any oscillatorsignal with success, no problem. But in faster markets, at first the 5 Min Chart is to long, next the cyclelength of the CCI`s is much to long for any useable signal in time, if the cyclelength differs and you get more chop and spikes and nice fakes, as usual a lot in the ES, nothing works no more. The same in very slow markets, the CCI is far to short in period.This system works only under very special conditions, the cyclelength has to meet the CCI 20 more or less and there must nice, rounded swings with no chop. If i read correct, the exit was a Buy/sell Trigger cross of 5 Min CCI ? If this is correct, this works definitiv not, the exits are much to late or much to early, depends on goals...did a very crude TS Simulation. The Cyclelength problem is the reason why indicatorsystems are not stable...in general I expect this to be a so called "random reinforcement" thing, Traders observed some behavior to work over a relative small sample of time, find some method working. But Mr. Market is just fooling, it`s just a random behavior for a limited time. Some random selected Chart ES with a crude version of system on...the thing with stop trading after stopped out outside yesterdays range is tricky, i simply don`t trade if price is outside Y Range >> best results, more worse if trading ouside Y Range. And i took no CCI-signal if New High/Low was more than 2 bars ago when crossing. May be i missed something with the exits, if things are not optimal (as most of the time...) the CCI/avg crossover as exit is to early or to late, not useable at all. Without any good idea/reason why the tradingday should not be a trendday, nor a choppy ugly day, or a complete flatliner, i would not trade this thing, next the stop one tick from LOD/HOD seems not a good idea to, sounds logical but everybody use it, stopfishing occurs often in the ES, not good idea. Michael
I see the same problem. I have actually implimented the WorkSpace in Tradestation 8.1 using TradeStation's datafeed. The CCI and CCIAvg for 5 minutes look similar to NPP's chart. But the CCI and CCIAvg value for 30 minutes doesn't look like NPP's chart at all. Here is an attachment of my screen shot for 3/31.
....some additional thougts about using indicators like CCI or any other... indicators work much better on longer timeframes as weekly/daily, cause the limited bandwidth of marketspeed/chop. Its the cyclelength problem why many experts looking for some solutions >> kaufmann`s AMA, jurik MA etc. especially on intraday timeframes you can NOT trade indicatorsignals mechanical, it needs a lot of experience to decide if the market fits the useability of the indicator, or to which % this is a good or bad signal. It`s far better to use Tick charts and not fixed time charts if using indicators, cause here is some adaption to the cyclelelength . In low activity market you get less Ticks/time and a slower/smoother Indi., in high activity you get more Ticks/time and a faster responsing Indi. But that`s far from testable, cause the amount of Ticks you receive varies from Datafeed/Broker, if using IB you get only ~ 1/10 of the real Ticks, a IB based Chart working fine ~25 Ticks/Bar is around 233Ticks/Bar on Esignal. The way I know to catch some ticks is to wait for a new low/high, the bar has to be a larger bar with a significant Volumespike, the volume larger any volume the last 1-2 hours, don`t wait for any indicators, don`t even wait for end of bar, look on a very short Tick chart to find a Reversal, take profits fast, use breakevenstop fast. And sorry, difficult to test..... Michael (Don`t believe nothing, test everything...)
have you tried creating your own 30 min CCI indicator to replicate Tradestation's? i can replicate the 30min CCI but the 10 period average of the 30min CCI does not match Tradestation's. if i calculate the average manually on a piece of paper it matches my indicator but it does not match Tradestation's.
would you please tell me how did you insert the 30 min CCI under the 5 min ES chart and the 5 min CCI on TS 8.1? thank you