OmniVest User Forums
Risk Control Requests (& other Account Settings)
Tiered Risk Control
2 replies, 1135 viewings
|
OmniVest User Forums
Risk Control Requests (& other Account Settings)
Tiered Risk Control
|
Last Activity 5/17/2016 11:06 AM 2 replies, 1135 viewings |
|
|
|
| Printer friendly version |
Jim Dean Elite   Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA  User Profile |
However my recent posts about the "EVG" (Equity/Volatility-Gap) stop are actually FOUNDATIONAL to many of the other more robust MMRC methods. So I really hope that one will get plugged in soon: http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4009 I mentioned briefly in that post that the EVG could be taken all the way down to the Strategy level but that imho at that level something different is needed. THE SIMPLE thing that could be done AT THIS POINT would be to allow the user to apply the EVG at the Strat level (in conjunction with the option of applying it at Portfolio &/or Accounts level with different settings). If this is done, then I believe you should offer it with the four inputs described here: http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4004 ... PLUS one more input: 5. dropdown box or radio buttons to use the EVG stop WITH the existing Strategy stops, or INSTEAD OF the existing strategy stops. ==== I'm calling this the "simple first step" since I think that down the line there is room in OVest for a SEPARATE TIER of selection options ... that is, allow the user to MIX AND MATCH different Strategies (ie System, Performance, Vote, Filter, Confirm) to define ENTRIES, and then optionally allow them to choose from a healthy list of alternative EXITS that would either override or parallel the native exits in the Strategies. Each Exit choice would have its own description, and statistics could be generated for it by using 2-3 "straw man" Entries: 1. an "ideal Pivot Point entry", much like NPP for exits, would be the best 2. a simple TREND entry system that generates lots of signals but is filtered to avoid bear markets and extended consolidation periods. 3. a simple SWING-TRADING entry system that generates signals during NON-trendy markets. With those three "straw-man entry" baselines, the various Exit options could be presented on Equity curves with statistics, etc to help the user select the one(s) to use for the "real" strategies. And of course once the Entry and Exit methods are chosen, a custom Equity curve could be created. This is clearly a "Future Enhancement" request, but I'm posting here in the Risk Control section since it is right at the CORE of risk control. ===== But, as I said earlier, a SIMPLE IMMEDIATE option would be to just offer the EVG stop as a replacement or companion to the native strategy stop ... ideally with the ability to see how the equity curve looks with it in force (along with any Portfolio and Account level EVG exits). Thanks | ||
| Jim Dean Elite Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
OVest currently has, or has "on the way", a very healthy structure upon which good position sizing and risk control can be done, at the "portfolio" and "account" level. The paradigm seems to be echoing that which a typical investment counselor might advise. Well and good. But since most OVest customers (for now at least) are Traders rather than Longterm investors, the portfolio and account level controls are insufficient for proper balancing and Risk management. This is since short term trades differ so much from one another - they "wiggle" more than a composite portfolio or account does. (One would hope :-) So, what's needed? Many things, some of which have already been described. 1. Fixed-max OR worst-case Trailing-stop control at the strategy level, using the EVG method. Pure Volatility will not be sufficient to cover all cases, in my experience. It needs to be a combo as I've described. I could give a bunch of examples but hopefully y'all will trust me on this one. 2. The trade size used needs to be related to the EVG-based Risk being taken on that trade. This Risk is NOT the total investment in the position, but rather the difference between the Current stock price and the EVG stop (pref a trailing stop). There are a few aspects to this - but the starting point is that the DOLLARS at risk for each trade should be CAPPED at some user-defined (small) percentage of Portfolio (not Account) overall (not available) capital. Typically this might be 1-2% max. 3. The trade size, for liquidity control, should be CAPPED at a small percentage of the average daily volume for that security - typically maybe 0.1% max. This is already in play but not afaik at the individual trade level. 4. The trade size should also be CAPPED at a certain max percent of Portfolio (not Account) Equity - this percent depends a lot on how many strats and trades are anticipated for that Portfolio, but 5-10% Max is about right for daily-bar basis. ... That's all pretty straightforward and is understood by many folks who practice careful Mandy management. But there is a lot more … 5. The EVG-risk (ie #2 above) should be SUMMED UP SEPARATELY for all longs and shorts in the Account (not the just the Portfolio). In a "neutral" market condition (whatever that is ;-) the sum of the Risk in Longs and Shorts should be equal. This is accomplished by auto-scaling the percentages used for sizing in #2, #3 and #4 proportionately (for all portfolios in the Account) to maintain the balance, moving forward - that is, those three percentages might be different for longs and shorts at any point in time, to "trim" the balance of summed Risk. 6. For non-neutral markets, the same #5 procedure is used but the target long/short ratio if EVG summed Risks (not equity in trade!) is altered to match the market direction. Only in string bull or strong bear markets (ie lingterm ones due to global political-economic climate) should the Trades all be allowed to be one type - and even then I prefer a 90/10 relationship. 7. The final "cash risk balance" component relates to the "craziness factor" that currently applies to the market. Craziness implies unpredictability. When the market gets like that, then the SUM OF THE SHORT+LONG percentages in #5 & #6 should be reduced from 100% to something less - approx 50% for really uncertain times. … ok that is enough for now … there is a whole 'nuther chapter about diversification. But this is the critical stuff. And btw - I would not count on "survey votes" to drive this decision - it is NOT well understood by novice or discretionary traders. This is something that people normally pay me to teach them. Some of the ideas are found in scattered books and articles but to my knowledge there is no single book that teaches this. [Edited by Jim Dean on 11/7/2012 5:33 AM] |
|
|
|
|
Administrator
Forum Moderator |
Registered User
Unregistered User |
 E-Mail this thread to a friend |
Toggle e-mail notification |
