OmniVest User Forums
General Discussion Topics
cost of trading
24 replies, 1508 viewings
|
OmniVest User Forums
General Discussion Topics
cost of trading
|
Last Activity 4/15/2019 7:43 PM 24 replies, 1508 viewings |
|
|
|
| Printer friendly version |
keith Member Posts: 23 Joined: 10/11/2012 Location: anaheim hills, ca  User Profile |
since then i have read nate silver's 'the signal and the noise'. he did a pretty extensive look at trading. he uses a 'dynamic momentum strategy' example of buying and holding versus buying and selling on signals. based on the transaction costs of .25 per share, the buy/sell method $10k initial investment would be worth only $1,100 ten years later. while i realize gxtrader is significantly cheaper, if we have to add in data costs, i'm not sure how that will impact our overall profits and/or losses. [Edited by keith on 2/1/2013 7:31 PM] | ||
| kmcintyre Elite    Posts: 890 Joined: 10/11/2012 Location: Portland, OR User Profile |
Many brokers have a minimum too. Unless your account is much larger than 10K your trade sizes will be so small that you will be paying much more than $0.25 cents. Take MBT with a $5 minimum. If you are trading 5 shares of a $100 stock (5% of a 10K account) you will be paying $1 per share commission. If you make $0.25 on the scalp, you lose $0.75 per share. Get a GXTrader account... (But I could be wrong... YMMV) Cheers! | ||
| Ed Downs Elite Posts: 645 Joined: 2/7/2007 Location: Austin, Texas User Profile |
Let me assure everyone that we are adding longer term Strategies that stay in trade longer and on average generate larger profits than the average RTM Strategy. The value of RTM Strategies is they take small profits out of the market very consistently, leading to equity growth over the long haul with low drawdowns. But they do produce a lot of trades and low transaction costs are important to capitalize on them. We will also very soon be releasing Real Time Strategies that will introduce a new dynamic into OmniVest that I think is going to be very powerful for generating consistent gains. But there again, low transactiono costs are going to be important. | ||
| DON SMITH Member Posts: 14 Joined: 10/11/2012 Location: CORNING NY User Profile |
Over a 5 year simulation (starting with $100000) @.005 the final equity is $3.86M and @.004 the final equity is $5.15M.Even a reduction to .0045 would be a great plus. Thanks Don S. | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
To save $1.3m as an example at 20 cents difference in brokerage/100 shares requires 6.5m trades in 5 years! John | ||
| Jim Dean Elite  Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
I believe that Nirvana Club members currently get 0.04 from GX | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
I realise that it's possible to build huge positions depending on the simulation settings and therefore commissions can increase, but it's very different to simulate trading 5K, 10K or 100K on a single stock, compared to doing it for real! It would be really helpful if the Settings page could be fixed for known errors, and an Excel export facility incorporated. John [Edited by John W on 2/2/2013 5:26 PM] | ||
| Ed Downs Elite Posts: 645 Joined: 2/7/2007 Location: Austin, Texas User Profile |
| ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
I can’t comment about settings issues raised by others, but if I just focus on two recent issues I have raised, the first may have been fixed, there hasn't been a recurrence http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4246 The second has not been fixed; I just tried this again now. http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4195 A general comment, if the developers fix an issue could you please ask them to have the issue moved to the resolved issues site. That way we know what has been fixed, it reduces the backlog of issues we provide information on, and if we have a recurrence of a resolved issue we can post accordingly. I think that would make it more efficient all round. John | ||
| Ed Downs Elite Posts: 645 Joined: 2/7/2007 Location: Austin, Texas User Profile |
| ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
| ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
[Edited by John W on 2/5/2013 11:41 PM]  Attached file : Unchanged Equity.png (149KB - 505 downloads) | ||
| Fred Gordon Legend Posts: 481 Joined: 10/11/2012 Location: Fayetteville, Ga User Profile |
Margin interest paid to GX for a multi million account which maintains a fully invested status is likely to be quite high. Is it's cost reflected in "final equity"? Thank you [Edited by Fred Gordon on 2/6/2013 4:17 PM] | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
Or if you have a broker that charges margin interest on a multi-million dollar fully invested account that trades hundreds of times a month my best suggestion is change brokers! John | ||
| Jim Dean Elite Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
I do not understand why OT and omniVest do not offer to model this - its not complicated. I also wish that OT and OV modeled slippage - I've provided complete algorithm: http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4292 Sure, I realize that both of these things would hurt the simulated performance of all nirvana strategies but its better to know how much hurt - rather than having to guess and worry. Nirvana please add these real-life factors, to both OT and omniVest, so that equity curve results do not continue to be inflated by some unknown amount. Thanks [Edited by Jim Dean on 2/7/2013 4:53 AM] | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
1. Margin rates for large accounts don't exist for actively traded large accounts 2. Slippage is a different issue With regard to slippage it's difficult to model realistically except in a limited way referred to below. Sometimes it happens, sometimes not. If my trade is large enough I can cause slippage. Maybe I can model that (e.g. my simple trading rule is if I trade greater than 3% of the volume traded in a day then that can move the market), but I can’t model what the other guy is doing. For instance if a half billion dollar hedge fund is an OV user and just happens to pick on one or even worse more than one of my strategies for its trades then there is potential for significant slippage even though I may have only put 0.01% of my portfolio on the line. That can't be modelled because the hedge fund may or may not ever use OV or my strategy, may or may not pick on the same stock at any point in time. If the hedge fund chooses a strategy that I'm not using and there are no other significant volumes in the stock then there won't be any slippage if I don't have a multi-any large number account. That’s when as slippage model can work because I'm modelling my impact and not figuring in any other players. But it’s not realistic as OV becomes more prevalent in the trading community. What’s needed is to model who else is going to use the same strategy as me at the same time as me and how many dollars they are willing to chuck in when I put my dollars in. I'm not sure that can be modelled, I suspect it's random and I'm not even sure it's any different to what's included in the open and close prices that are posted daily? John | ||
| Jim Dean Elite Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
Yes of course slippage is very different than margin rates! Their commonality is that they are both "costs of trading" which is the topic of this thread. AND that neither is modelled by OVest. Sometimes margin rates matter a lot, sometimes not. Sometimes slippage matters a lot, sometimes not. Thus the need for intelligent modeling. Before you reject the idea of modeling slippage out of hand, please review and consider the algorithm that I provided at the link I included I my prior post. It is not perfect but it does account for the two major contributing factors to slippage - liquidity and volatility. http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4292 It may be that in your experience slippage does not matter much - that may relate to the type of trading you do. But in my experience, it can and does matter a lot - esp with short-duration trades such as RTM strategies. There is NO HARM in providing the modeling capability - it can be turned off or tweaked to suit the beliefs and trading style of the individual user. But if the capability to model slippage and margin rates (or any other quantifiable real cost in trading) is left out of the engine, then the net effect indisputably is to positively bias the simulation results to some degree or another. Please pardon my assertiveness - it is not focused on you personally. I feel very strongly about this need. [Edited by Jim Dean on 2/7/2013 6:55 AM] | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
Yes I agree that both margin rates and slippage are costs of trading. I agree intelligent modelling is a wonderful tool to have available. I do not reject the notion of modelling slippage. Slippage is a fact of life, and I have previously agreed by separate email that your model for slippage is fine with me. But that model is all about your trading account causing slippage. The issue that needs further discussion IMHO is how to model slippage if we are not the party that is likely to cause the slippage to occur. So I'm curious to know how you would simulate the slippage caused by my hedge fund example where you don't know if they are using OV, you don't know if they are going to trade your strategy, you don't know the size of their bet, but you do know you are just trading a small parcel to ensure you don't cause slippage? I'm indicating the slippage issue that matters is not the little guys but it’s with the other bigger guys using OV, how can you model what their behaviour is likely to do to your account except to say that past slippage data from that bigger guys trading is already in the open and close prices. Please pardon my assertiveness also, it’s also not personal, often the typed word is a very poor substitute compared to spoken dialogue and can often cause misunderstandings. John [Edited by John W on 2/7/2013 7:36 AM] | ||
| Jim Dean Elite Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
I think I see the reason for this ... we are looking at it from different perspectives. Your perspective seems to be mainly on the proximate causes of slippage, and how my trading does or does not relate to those causes. The "causes" you've mentioned include hedge funds and presumably the future collection of OVest users, all entering and exiting at similar times. Some would ascribe it to the nefarious machinations of brokers or other Level III players. My perspective is that all that stuff is beyond my control or my ability ever to discern, therefore it would be a waste of my time to try to build a slippage model that somehow anticipated or responded to presumed actions by bigger players. So, I look at it simply as force majure, and approach it from a different tack ... I try to model the EFFECT rather than the CAUSE. I believe that the slippage EFFECT is significant to most active traders, and less to to longer term investors. It's my belief that Nirvana's market is the former. So ... I study the effects, to search out measurable and rational factors which might be used to estimate it. The estimates are of course "averaged" ones ... not particularly good for zoomed-in focus on a single upcoming trade prediction, but VERY useful for longer term simulations of equity curves, where statistical rules can legitimately be applied. Modelling such things is primarily limited to the data available to us. Historical bid/ask data is not available, so we need to look for "rangy" price info that logically would be a reflection of bid/ask spreads. My own studies of this, comparing to bid/ask realtime info, have shown a useful correlation to ATR measurements. There are other possible measurements which could be factored in such as the ratio of body size to bar-range (small ratios often imply big spreads) ... but for the sake of maintaining a balance between precision of calculations versus the nature of the solution needed, I limit the calc to an ATR-multiple basis. Also, it's obvious to anyone who has done significant trading that the likelihood and degree of slippage is driven by the liquidity of the instrument. This is usually relative to the size of the trade being presented ... if the trade is tiny versus the overall volume, the slippage is much less than if the trade is a significant fraction of the volume. So, over the years I've used an approach that melds these two pretty-well-defensible relationships with observation-based scaling factors. It seems to work pretty well. Of course it may well be that the scaling factors need to be somewhat different for entirely different types of instruments ... funds, futures, options, stocks all act differently for many reasons. I have SIMPLIFIED the modelling that I use into the tabular example that I provided at the linked thread. Nutshell: rather than beat my head against the wall attempting to solve the mystery of "why", I've sought a more beneficial route of trying to model "how much". I hope this clarifies things. It is the mark of an educated man, and a proof of his culture, that in every problem he addresses he looks for only so much precision as its NATURE permits and its SOLUTION requires" (Aristotle) [Edited by Jim Dean on 2/7/2013 2:45 PM] | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
In another thread I mentioned that in the Russell 1000 I ran an OmniScan to find out how many stocks on average traded less than 1000K, 2000K and 3000K per day [e.g. Avg(C,14)*Avg(V,14) < 1000000]. I’ve run that again today and there were 44 in the less than 1000K category, 156 in the less than 2000K category, and 258 in the less than 3000K category. Using my belief that 3% is the maximum to trade before the price starts to move then 3% of these 3 categories is 30K, 60K and 90K respectively. So let’s take the higher example, the 3000K/day traded stock. At the moment there are 400+ members in this forum. Let’s imagine a really popular strategy is followed by 50% or 200 OV users. Let’s imagine the average account size is 50K, and that 10% or 5K is the normal trade size. So at a particular time 200 OV users place 5K on stock ABC at market open (or the next bar open on a real time strategy), and sell it another few days (or bars) later. This is more than 3% of the daily traded volume to be placed by a group of OV users acting as one at one point in time. My belief is that this will cause a gap in the price for the 258 stocks in the less than 3000K category, bigger gaps in the 156 stocks in the subset trading less than 2000K and even bigger gaps in the smallest 45. My other belief is there will be the fall off in performance of the particular OV strategy in play in the smaller symbols lists because now the group of OV users trading this strategy as one will experience a gap up when they buy and a gap down when they sell, lessening returns over time. It’s not hard to imagine that some people trading OV may put on larger trades than the $5K example, that the number of OV users could exceed 1000 or even 10,000 (one competitor claims more than 65,000 members!), and even hedge funds and other big players may be attracted to OV. The 100K traded on single stock could grow massively which will cause many of us to re-evaluate our returns on smaller symbol lists. If the number of users using OV jumped to 1000 then the cut off in my example is 6500K and only the top 500 stocks have the possibility they will avoid some form of slippage caused by OV users acting together. At 10000 users only the top 62 stocks escape the possibility of OV induced slippage if we all act together as one! In summary I'm saying that it may be a waste of time building a slippage model for a single account. There won’t be observable slippage caused by that single account, just gaps and future performance issues in smaller lists in particular. The way that I’m hoping Nirvana plans to combat this is to consider adding many more strategies and introduce other data sets such as futures and options, even FX and different time frames, even other country lists, limit orders and other Elite strategies offered by others to keep growing the universe of available alternatives so that as the number of members in OV grows the risk of opening gaps is kept manageable so as not to damage the future performance of OV. Nirvana could also introduce random small purchases throughout the day in the same manner as the big players in the market, and also could limit membership or the number of strategies to be traded. Ed has already mentioned some of these alternatives so I’m firmly in the camp that the guys in the control room at Nirvana have a path mapped out, and over time the building blocks will come together. I apologise for coming at this from a different perspective but I do believe that slippage models for a single account may not be relevant in the OV scenario of the future. John | ||
| Jim Dean Elite Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
There are a lot of "let's assumes" in that scenario. And basically you are building a case that EMPHASIZES that "slippage" will be a HUGE impact, even though it's not really "slippage" technically ... it's due to the rapid simultaneous placement of many orders by many customers on the same symbol at very close to the same time. Yes, Nirvana might choose to act like a "fund manager" and "scale in" the various trades ... but that would be HUGELY UNFAIR to the people at the end of the list ... and HUGELY BENEFICIAL to the people at the start of the list. If Nirvana chose to do this ... I would never use OVest canned systems to trade with, ever ... and I would advise those who look to me with trust for recommendations similarly. But, regardless of what I would do, the point is that for one reason or t'other (please see my prior post re looking for why's rather than how-much's) ... the slippage (ie the unexpected discrepancy in price) would be HUGE if what you suggest came in to play. So, what's an OVest'er to do? FIRST OF ALL, they should "pester" Nirvana to provide SLIPPAGE MODELLING in the OVest Equity curves, so that whatever the relative magnitude of the slippage IS, it is ACCOUNTED FOR. Even if the estimation is inadequately low (which I believe is your point), that's better than NUTHIN! And if the method I proposed is implemented, the USER can decide to model the "big whammy" impact if they see that happening, and if they choose to use the canned Nirvana lists and canned Nirvana strategies. That is ... give the choice TO THE USER to model real-life impacts as a part of the Simulation ... SO THAT they can make intelligent decisions on what strat's to use. That is, after all, one of the top 2 or 3 selling points of OmniVest. ------------ However, there is another way out of the problem that you mentioned. It has to do with another area that I've repeatedly asked for feedback about - see (in inappropriately-named thread): http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4092 That is ... if an OVest user chooses to implement CUSTOM strategies via an Elite-Trader interface to OmniVest, then it is much less likely that the "pileup" of orders will occur, even if that ET-strat has some "subscribers". ALTERNATIVELY, or additionally, if the OVest user is allowed to specify their own Symbol list that intentionally DIFFERS from the symbol list that is "canned" with a "canned" Nirvana strategy, THAT would potentially reduce the pileup-problem. ---------------- Those are just suggestions. The point is ... slippage and margin costs are REAL. And they can be very significant. The PROBLEM is, the Equity curves produced by OVest IGNORE both of those effects when it would be programmatically not terribly difficult to include them with user-configurable parameters. All I am asking for is the ability to do that. If someone else thinks it is unnecessary or whatever, fine ... DONT USE IT. But please, please do provide the capability to those of us who DO believe it is essential. ------------- Again ... summary ... this thread, and my requests, have NOTHING to do with how the Trade Processor works, or what overall impact OVest itself might have on the market. THIS THREAD IS ABOUT MODELLING REAL TRADING COST EFFECTS IN OVEST SIMULATION MODE. Please let's not lose sight of that. Please, Nirvana ... include these modelling capabilities for those of us who will not trust the Equity curves otherwise. There. I said it. No criticism intended. But for me, that's the bottom line. | ||
| Steve2 Elite Posts: 750 Joined: 10/11/2012 Location: Annapolis, MD User Profile |
| ||
| Jim Dean Elite Posts: 1059 Joined: 10/11/2012 Location: L'ville, GA User Profile |
BUT ... that would take a lot of time to gather and no small amount of fancy-math to implement. I would hope that the simple and understandable paradigm that I've offered would be usable much sooner than that. Thanks for the suggestion! | ||
| John W Elite Posts: 654 Joined: 10/11/2012 Location: Sydney, NSW, Australia User Profile |
A couple of points, perhaps in coming at slippage from different ways we are getting to the heart of the issue. This is why forums such as this one provided by Nirvana re so valuable. When I talked about Nirvana acting like a fund manager I envisaged Nirvana placing random small entries or even intelligent opportunistic entries depending on buy/sell volumes to scale in everybody at the same time, so all users including Nirvana 'insiders' were neither advantaged nor disadvantaged. We all scale in together at the same times throughout the day to keep it fair for all. That’s just ethical and I expect that’s the way that Nirvana would do it! I think the points made in the thread you refer to at http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=4092 is something I’ve commented on too and is exactly what is needed for Elite strategies, and yes I believe that will have a positive beneficial impact and reduce the risk of gaps and lowered performance. Also user added symbol list as briefly discussed in http://www.omnitrader.com/currentclients/omnivestforum/thread-view.asp?threadid=3956 would spread the trades wider. But let's get to the main topic.. I think that Nirvana will have the statistics to know how many $ are being put into each strategy every day. To build on your concept perhaps Nirvana could use your suggested model as an input but rather than you or I guessing how many dollars are being traded by the sum of OV users interested in that strategy on a particular symbol perhaps Nirvana could pre populate your slippage model with its knowledge of the likely slippage for each symbol based on its continuing and growing extensive knowledge of $ traded. User could override the pre-populated slippage of course. Needs a bit of thought but this is Nirvana’s and your area of expertise Jim, not mine, so perhaps there is a way. Using real OV data of the $ traded by OV by strategy and symbol allows realistic simulation of the changing nature of our equity curves as time and the number of accounts and account sizes of the users rolls on.... P.S. While I was writing this I notice that you and Steve2 have similar more recent thoughts as well. It looks like different perspectives can often lead to better understanding and shared agreement! John | ||
| Steve2 Elite Posts: 750 Joined: 10/11/2012 Location: Annapolis, MD User Profile |
Jim has made a request for this in the OV enhancements thread. I would suggest that folks read that thread and weigh in (pro or con) so Nirvana can gauge our interest. |
|
|
|
|
Administrator
Forum Moderator |
Registered User
Unregistered User |
 E-Mail this thread to a friend(un)/Freeze thread | |
Toggle e-mail notification |
