OmniTrader Upgrade Forums
ATM
New Macro + Micro Method
100 replies, 16462 viewings
|
OmniTrader Upgrade Forums
ATM
New Macro + Micro Method
|
Last Activity 4/21/2021 9:48 AM 100 replies, 16462 viewings |
|
|
| Printer friendly version |
Jim Dean Sage   Posts: 3022 Joined: 9/21/2006 Location: L'ville, GA  User Profile |
That’s different than what Mark is describing (an overlap of MS's) … "overlap" does *not* imply that a formula for MS which currently doesn’t work if getclose($Group) or getclose(Symbol) is used in it, *will* be acceptable. But I certainly hope it will be. My viewpoint on MS analysis is that the tiered formula capability is crucial, since no one index Symbol is a good avatar for a big FL of trade-Symbols [Edited by Jim Dean on 5/2/2018 9:44 AM] | ||
| gary Member Posts: 16 Joined: 12/13/2003 Location: Houston, TX User Profile |
I also ran portsim over the same time period to see if maybe autotrade was not working right. But got very similar results shown below: I used the standard ATM settings from Mark with the exception of allocation and strategies. Changed the allocation to use the optimised values found in an earlier experiment. The autotrade settings are as follows: I can supply info on trades etc. if someone is interested. I am wondering if these results are just dealing with a tough market or if I have done something wrong. If others have similar results, I guess it is just the tough market. Since the current market is very volatile I was hoping the RTM strategies would do well. Maybe the turns were just too sharp. It took some heavy losses. see below position history: [Edited by gary on 5/7/2018 10:45 PM]  Attached file : 2018-05-07_equity curve.jpg (240KB - 1445 downloads) Attached file : 2018-05-07_portsim run.jpg (117KB - 1432 downloads) Attached file : autotrade settings.jpg (59KB - 1385 downloads) Attached file : 2018-05-07_position history.jpg (616KB - 1380 downloads) | ||
| mholstius Veteran   Posts: 175 Joined: 1/13/2017 User Profile |
I haven't been using the paper trade capability in ATM, and I'll agree that the market has been tough lately. But, when I look at my ATM trades I only show 36 trades from 3/29 - 5/4 vs your 73. Could you do a "View Trades" export from ATM port sim (preferably to excel) and attach it? Also, if you don't mind showing it, it would help a lot if you could also post the details scrolling down on the right side of your port sim page (that'll show all your settings); It looks like we do share at least one thing in common - the uncanny ability to start trading a system just before a drawdown... ;-) Mark [Edited by mholstius on 5/8/2018 5:16 AM] Attached file : Port Sim Details.png (633KB - 1342 downloads) | ||
| Barry Cohen Sage  Posts: 6338 Joined: 1/19/2004 User Profile |
| ||
| gary Member Posts: 16 Joined: 12/13/2003 Location: Houston, TX User Profile |
[Edited by gary on 5/8/2018 1:05 PM] Attached file : portsim1.jpg (248KB - 1335 downloads) Attached file : portsim2.jpg (387KB - 1312 downloads) | ||
| gary Member Posts: 16 Joined: 12/13/2003 Location: Houston, TX User Profile |
here is the portsim run with only forming signals enabled. the equity curve shape is a little different and the results are a little better but not by a lot. [Edited by gary on 5/8/2018 1:15 PM] Attached file : portsim-onlyForming.jpg (146KB - 1283 downloads) | ||
| Barry Cohen Sage Posts: 6338 Joined: 1/19/2004 User Profile |
| ||
| gary Member Posts: 16 Joined: 12/13/2003 Location: Houston, TX User Profile |
btw, since I am using optimized allocation parameters (and I suspect you are not), some difference between our results should be there. Hopefully, the optimized parameters would make mine a little better than without. But still not apples to apples since I substituted VBX-3 for RTM-7. thanks barry | ||
| Barry Cohen Sage Posts: 6338 Joined: 1/19/2004 User Profile |
| ||
| John W Regular Posts: 87 Joined: 8/1/2011 Location: Sydney, NSW, Australia User Profile |
“Max Long% and Min Long% automatically establish the value of Max Short %, since it will be the inverse of the Max Long%. That is, if Max Long% is 80 than Max Short% is 100-80 = 20%.” To follow that logic then if Max Long% is set to 100 then the inverse Max Short% should be 100-100=0. Therefore no short trades if Max Long%=100. In the ATM Macro and Micro Method the Allocation for EVERY Market State is Max Longs % 100, therefore no short trades should be possible (100-100=0) in the Macro and Micro Method. Yet short trades occur. It appears that the manual is incorrect, and it’s also unclear how the ATM module chooses the maximum percentage allocated to long or short. In Omnivest there is a very good system in the Trade Sizing Section that allows the user to choose: Max Long Allocation % (e.g. anything from 0-100) Max Short Allocation % (e.g. anything from 0-100) I believe this OV approach provides great clarity and flexibility and the sum of max long and short does not have to be 100 (e.g. 100,100; 0,50; 50,50; 100,50; 80,40), and adoption of a similar approach in ATM is recommended! | ||
| Barry Cohen Sage Posts: 6338 Joined: 1/19/2004 User Profile |
If you wanted no shorts defined in the allocation percent parameters, you would set both the Max Longs % and Min Longs % to 100. By setting Max Longs % to 100 and Min Longs % to 0, basically anything goes - longs & shorts. In that case the longs/shorts are solely defined by the Max Longs & Max Shorts parameters. | ||
| John W Regular Posts: 87 Joined: 8/1/2011 Location: Sydney, NSW, Australia User Profile |
So to define a market state with 0% Long, 5% Short, I'd express that as Max Longs % 0, and Min Long % 95? To have minimums exceed maximums does seem counter intuitive to me, perhaps for others also, I encourage you to expand the description in the manual. BTW ATM is just AMAZING, WELL DONE Mark Holstius and Nirvana! | ||
| Barry Cohen Sage Posts: 6338 Joined: 1/19/2004 User Profile |
| ||
| FG Member Posts: 39 Joined: 1/8/2004 Location: Atlanta, Ga User Profile |
Thank you, Fred Gordon | ||
| Jim Dean Sage Posts: 3022 Joined: 9/21/2006 Location: L'ville, GA User Profile |
Imho the important Q is how often N updates the “All Optionable Stocks” general list, from which other subsets are derived. Barry / Cose would know that. [Edited by Jim Dean on 6/4/2018 10:20 AM] | ||
| Barry Cohen Sage Posts: 6338 Joined: 1/19/2004 User Profile |
| ||
| Jim Dean Sage Posts: 3022 Joined: 9/21/2006 Location: L'ville, GA User Profile |
Is there some way to know, historically, if a given stock was optionable on a particular bar (ie fundamentals or some other means). If not, could you get that added to the fundamentals list? Thanks. If that is available, then it's possible to create a "historical tracking" variant of Mark's list. | ||
| FG Member Posts: 39 Joined: 1/8/2004 Location: Atlanta, Ga User Profile |
| ||
| LSJ Legend  Posts: 515 Joined: 8/17/2006 Location: Citrus Springs, FL User Profile |
I am about to start experimenting and would like any ideas or comments. [Edited by LSJ on 6/14/2018 11:35 AM] | ||
| mholstius Veteran Posts: 175 Joined: 1/13/2017 User Profile |
I haven't tried it, but look forward to hearing what you find. Mark | ||
| Jim Dean Sage Posts: 3022 Joined: 9/21/2006 Location: L'ville, GA User Profile |
In order for backtests to be valid, you need to restrict the scope of possibilities to the dozen or so historical fundamentals that N provides - you can find them in the function list in the OLang or Formula Builder editor panel. Here is a snapshot: [Edited by Jim Dean on 6/14/2018 11:49 AM] Attached file : Historical Fundamentals.png (128KB - 741 downloads) | ||
| LSJ Legend Posts: 515 Joined: 8/17/2006 Location: Citrus Springs, FL User Profile |
I wasn't thinking of stock fundamentals as much but of market internals. There is an easy list available in Omnidata and Easydata. The attached are the internals that have current data. I don't think all of these are useful but it is what there is to pick from. I have charted many of these and am considering how to use them. In some cases it seems a moving average crossover or the data crossing a moving average corresponds well to movement on the SPY chart. Anyway, a lot of sorting and testing will be needed to get this down to useful in market state and if it is useful. I'm thinking ATM2 may also expand that use by allowing differences in NYSE and Nasdaq, among other things. [Edited by LSJ on 6/14/2018 12:14 PM] Attached file : MktInt.png (24KB - 739 downloads) | ||
| Jim Dean Sage Posts: 3022 Joined: 9/21/2006 Location: L'ville, GA User Profile |
If the data you use in ANY part of the method is "static" ... such as using a locked-in Symbol list that actually does change over time, then the backtesting will be erroneous and the resulting tuned Method will be unlikely to perform as you expect in the future. Just a word to the wise. Unfortunately, since the majority of the market internals are calculated across many symbols to arrive at a single value, OLang cannot be used to approximate them. It would be possible for N to do it themselves ... but undoubtedly they'd require (chicken and egg) proof ahead of time that it has benefit. Sigh. [Edited by Jim Dean on 6/14/2018 2:02 PM] | ||
| LSJ Legend Posts: 515 Joined: 8/17/2006 Location: Citrus Springs, FL User Profile |
I'm not sure I follow you on that. Generally speaking I am contemplating using market internals in ATM states or filters. I don't understand what you mean by valid only for HRE and "static". I have historical data with daily change so could I not use a moving average or some other derivation from the data to define a filter, e.g.? I appreciate your experience here. (I am leaving for a few hours but will be back.) | ||
| Jim Dean Sage Posts: 3022 Joined: 9/21/2006 Location: L'ville, GA User Profile |
======== however ======== The backtesting issue IS important, for "static Lists" (ie symbols associated by you, with a particular market state) ... or if you're not using ATM, then any static Focus List based on some kind of external-to-OT filtering/selection procedure. For example, Eric Severance gave a short presentation at the Bash regarding the value of using Symbol Ranking based on the a well-known publication's "Growth" list of 50 symbols (name is made up since the source of the info is strictly copyrighted). What wasn't stated (as I recall), is that this also presumed the FL symbols are also *filtered* by whether or not they are on the Growth list. He and MarkH nicely demonstrated that using the Growth-ranked order in lieu of the default "alphabetic" order for selection of symbols that have signals on the same day, provides a definable edge. N was convinced, and they plan to incorporate that ability in OVest ... that is, if you upload a List, it won't auto-alphabetize it and lose the sequence of the rankings. This will be a BIG plus in OVest, I believe. OVest of course supports uploading Historical Lists ... that is, huge text files with lots of strung-together "sets" of symbols, for sequential dates in the past. For example, that "Growth" list comes out once a week. Eric has painstakingly gathered a lot of past weeks to create the file (which btw he cannot "give out" to others due to *very* strict copyright protection - they can and do take people to court about it). Aaaanyways ... this means that OVest has the ability to HISTORICALLY TEST the value of that Growth List's rankings (and the symbols on the list, for that matter), to demo how much going that route improves things. Hurrah, OVest! Unfortunately, N long ago decided not to support Historical Lists for OT/VT ... just too messy for them to deal with, I guess. Fortunately, I'm just about finished development of a comprehensive, robust tool that *will* fully support Historical Symbol Lists in OT/VT. Contact me privately for more info. So, in OT/VT, when your strategy or method is "geared" to the use of certain symbols for a given Market State, or certain symbols in the Focus List, and if that symbol-list is defined by: 1. an external algorithm or entity, such as a guru list or ETF 2. a periodic reshuffling of what symbols are on the list, and/or their order ... then, in that case, since OT (natively) does not provide a way for you to vary the makeup &/or order of the symbols on the list from week to week (or some other frequency) ... that becomes untenable for valid/useful Analysis. With that in mind ... Basic presumption for this all to matter: You are going to use optimization, or formal backtesting via Analysis/PortSim stats or via SW ... to decide what Method-variant works best. I think that virtually everyone does one or more of those things, but if you are the exception, then this "advisory" doesn't apply. If you are doing one of those "pick and choose based on statistics" methods mentioned above, then it's likely that (for daily bars) you're looking back at least a year and possibly more, at some point along the way. Remember ... unless you specify otherwise in the Strategy Test Settings, OT will run the stat's for all the bars you have loaded at the time. So, that's the "time window". Now ... if the List of Symbols you are using is the "most recent" (ie just before the HRE) state ... and if you run a 3-year (or whatever) backtest on that "static list" setup ... then there are two problems that arise: 1. in the recent past, the fact that you have future knowledge of the "successful symbols" builds in a crystal-ball rosy-glow bias. Ungood. 2. in the majority of the past before that, presumably the "HRE at the time" list of symbols has changed frequently. In fact, some of "today's" symbols might not even have existed at some point in your test window. NOTE: this applies even to basic Indexes such as the SP100, DJ-30, Russell-1k, etc ... all of which are updated typically once a year or more. Using the current list BIASES long backtests to show better results, since those outside agencies (SP, DJ, Rus) go through a process of selecting "strong" stocks. CONCLUSION: Static Symbol lists based on outside sources that change over time create ERRORS in the analysis results. The more frequently the outside source changes the Symbols in the list, the greater the error will be. And that's why I've spent a huge amount of time solving the problem, so Historical "sequential" external-list changes can be properly modeled in OT. [Edited by Jim Dean on 6/14/2018 2:03 PM] |
|
|
|
|
Administrator
Forum Moderator |
Registered User
Unregistered User |
 E-Mail this thread to a friend |
Toggle e-mail notification |
