With the ACD trade...not just one trade. Had about 10 of these in the past few days and about 75% win ratio with good R:R!
Just for kicks here is NEM with the noise level plotted for today. BTW I've been looking at VWAP as a filter and it seems to be very useful, as it should considering how many high volume players look at it.
I think this is more interesting. The first 2 lines close together are the 95% confidence interval of the average of the noise level over the last year of trading. The next line up is 1 standard deviation. The one above that is 2 standard deviations. That means that those bands will contain 95% of all noise values observed over the 1 year period. BTW that is an awful nice pullback to the 2std breakout point. It would make Al Brooks happy. Not to mention the iii entry that follows a couple bars later.
I also ran a test comparing pivot range size to the days range. There was absolutely no predictive value. It has an R value of 0.006 which means it couldn't be any closer to random if you made up the data.
Sorry for all the posts but got one more. I forgot earlier to plot the down values too. You can see the difference in plotting off the open vs the open range. The A value is way down somewhere off the chart, while calculating this way gave us a bullish bias pretty early in the day as all price action below the down values was overwhelmingly bullish. It kind of resolves that problem of not wanting to trade products with big opening ranges. I personally think the size of the opening range is not really predictive of the rest of the days action. It seems like way too many random factors influence the size of the first bar to really consider it meaningful.
Good to see you are doing some testing! What program are you using? Its redundant to test it to the prior days range because the pivot range is dependent upon the range of the prior day and they are essentially one and the same! I don't know why Fisher mentions to compare it to the previous day lol If you have the ability, test out the size of the pivot range relative to the 10 day ATR. If we could get a ratio that would be great. Also if the days range is smaller than the prior day's rage. This would just make it a NR day, but there may be some predictive value. Anything else you have looked at?
I know what you mean! some crazy moves happen right off of the open, so using the OR does miss some of those. Also I don't mind the posts, keep going...would love to see what else you are looking at.
I'm using gnumeric. Its a free spreadsheet program with a lot more stat capabilities than excel. I think you are wrong on the pivot thing. Pivot range = the regular pivot point + and - the difference between the pp and the midpoint of the day. Thats all I've done so far. I'm not even going to use any of this right now. I'm trading on a daily time frame and taking Al Brooks style entries after getting a monthly signal based on the levels calculated the way I plotted on that chart. (AUD is already confirming a 2std monthly breakdown btw). I mostly just realized I missed doing stuff like this and it sounded like fun today lol. The ideas you mentioned can definintely be done, but I think I'm all computered out for now. The program isn't too hard to learn if you feel like checking it out.
hmmm I looked back at the formula and you are correct. Don't know what the size of the pivot range represents in terms of a market behavior. Looking at the range of the day for range contraction makes sense, oh well they work great for framing a bias!
So haven't posted here in a bit and thought I would provide an update. Lately I have been running a strategy that I put together based off of ACD. Essentially forward testing in real time and tweaking and adjusting according to real world results. Had been backtesting this and it showed alot of promise and it was based on sound premises. Been trading with relatively small size and will be sizing up more next week as I am satisfied with my testing so far. These are the results using this approach for the past couple weeks: 209.16 96.13 201.86 34.46 (170.95) 47.49 (41.9) 160.58 152.14 (93.45) 45 Some of the execution and conditions will be changed next week after noticing some things from trading this approach for past couple weeks! Been reading up on some stat arb/pair trading stuff for the past month or so and looking to add that along somewhere in the future. Just have alot of work to do with that in order to have something that is consistent and can be executed properly!