| Vyrd |
|
Gold User, Member, TeleChart
|
| Registered User |
|
|
|
|
| Unsure |
|
| Tuesday, March 29, 2005 |
| Friday, July 16, 2010 3:54:26 PM |
56
[0.02% of all post / 0.01 posts per day] |
|
I have two ETF trades that never got filled despite sitting at the bid price all day. But that would seem to indicate that sellers don't want to sell below their ask price, not that they are looking toward the exits. Still, I have been expecting a correction for awhile.
I fired my useless broker yesterday, will be using one of two advisors recommeded by the company I am in the process of moving to. Even their historical performance on the sample portfolio is better, averaging 5% over past 3 yrs at one vs. -1.4% at the old one (slightly different time periods though). The real performance of the former (a real portfolio but not necessarily identical to the proposed sample porfolio) is a good deal lower though, -2.6%, reflecting slippage and delays or imperfections in their asset allocating "timing" model.
The one I will probably use has an ACTUAL -1.4% 3-yr average in their moderately aggressive portfolio, net of fees. They've all done pretty badly recently, LOL.
|
QUOTE (ben2k9) I like the simpler strategies that don't require scaling in.
Whan I was checking the advertised results on Connors' model portfolio page earlier in the year I found that the scaling in makes a huge difference in the returns of his system. The double commissions (buy x 3, sell all at once) will eat up some of that but probably not much.
|
QUOTE (Apsll) Vyrd, How are you doing?
OK, thanks for asking.
QUOTEThanks for the input. You are correct that a blanket test result for the entire period that you mention is quit bad; however there were three time frames (Short) that worked well on the SPX from November 2007 to March 2009 and one time frame (long) that worked well from early April to late June 2009. Of course as I have mention many times "when price is stalling or moving sideways then the system is rendered impotent"
I know that looking back is always a much better view of these things so I am not going to make much of my findings, but thanks for your efforts and interest...
I was really looking more carefully at a system I had revisited when intothetrade suggested using a system that works well on SPY and other mutual-fund-like ETFs. Unfortunately the EMA-based system isn't an all-weather system. Just having a system that can tell you when to use a moving average system successfully should be quite a lucrative accomplishment.
The CCI system works well throughout its history (as far as I have gone into it in detail, I find programming in Stata tedious compared to SAS, so I pretty much just use the canned program I wrote),
but the losers are equal in magnitude to the winners, even though they are 1/3 to 1/4 of the trades. It's probably not a tradeable system. Also a lot of the potential peak profit is lost, I can't seem to codify in rules how to successfully take the majority of these profits.
|
QUOTE (intothetrade) i actually posted larry connors strategies somewhere here on the board i believe
His 5 SMA / 2 RSI system works surprisingly well, that was easy to program in TFilter. I should look at the damn book, I paid enough for it ($45 w/shipping).
I have been somewhat soured on Connors and Trading Markets after they overbilled me twice. Even though I received a credit back to my account when I brought it to their attention, I wonder how often that happens and doesn't get noticed.
|
QUOTE (davidjohnhall)
I have seen Wealthlab used many times as well. I use Strategy Desk from TD Ameritrade. I have also used AmiBroker (which I believe Booker uses). There is also TC Companion which works pretty seamlessly with Telechart and tests PCFs -- there a great yahoo message board associated with it if you want more information.
I use Technifilter. It has a powerful language which allows multi-line formulas, so the period of an indicator can be the result of or modified by a prior formula, for example, or you can just simplify some of the horrendous PCF formulas I have seen by programming them in manageable sections; although I am sure WealthLab and TC Companion have many more nice features. TF doesn't seem that well supported anymore (for awhile it was off the market) and has no community that I am aware of. Some of its security-level backtest statistics don't always look right to me (not consistently, just sometimes), so I always output the results to an ascii file and anayze them in Stata. It's easier to look at them on a trade-by-trade basis that way anyhow.
I upgraded to Metastock 9 or 10 (I had owned it back when it was a Windows 3.1 product and they gave me the regular upgrade price! great customer service by Equis), but I never really took the time to relearn it, especially after I had a fatal HD crash a few months after installing it. Same for Supercharts, which I bought on ebay or via usenet and never installed.
Has anyone ever tried out the ART trading software? It won't work with EOD data so I have been reluctant to get Signal (even though they will also give a one-month trial, its just a great deal of trouble). The software is pricey, I bought the book on Amazon. Considering it is just a long advertisement for the software, one wonders why they don't give it away of the software is that great. I probably answered my own question.
|
QUOTE (Apsll) I remember that the method had somthing to do with moving averages. Of course in these markets I would have no use for the system.
I was looking at an EMA crossover system that tested a lot better on SPY and other large stock index ETFs than the CCI one, and then I decided to look at just the results since the 2007 peak. They were 5 to 1 losers with large (median of -35%) losses.
The CCI system backtested well on individual stocks in the recent time period, but on SPY etc the distribution of results has a long negative tail, so the median is around 10% but the mean is 2%.
|
QUOTE (tomson10)
Well actually there are trading systems that do give consistant results better than 70%. Larry Conners has published a book that has 7 systems that has an accuracy rate between 82.8% to over 93.9%. No money management skills needed. Just simple rules to follow.
Is that his High-Probability ETF Trading book (which I purchased)? The percent return per trade is in the 1-2% range, add commissions and slippage and I doubt you would make much money. They might be good starting points for system design though.
|
QUOTE (intothetrade) eh so you had practically your entire account in gold?
We weren't able to buy gold/PM funds in our retirement accts at work for a long time. They have been put back into the Vgd and Fid selections as of some time ago, but having been burned in gold funds before I have stayed out so far. It takes big brass ones to go all in like that, good luck to signaltap.
|
I get discouraged with trading and then don't do anything for months at a time. I was using Telechart to make allocations in my 403B with some success (8-18% per qtr a lot of the time) but I took my eye off the ball when two members of my family died within two months of each other and lost several years worth of gains in a few months last year.
Too bad about your aquaintance at Merrill Lynch. That's why I am here, to learn so I can do it myself without losing money or putting up with condescending lectures about market timing from overpaid gnomes (there's a hint at to the investment firm).
Since I posted my TSF system results back in April I am down around 7% in my trading acct, not good especially compared to the market's gains. I have changed it a lot since then, but's just hypothetical until I start making money with it.
I did have a one-day gain of 7% on RODM last week, but this was after a 70% runup since I got a signal (I was still tinkering with my system and wasn't doing my daily runs as I should), and then it was up 4-5% the day after I sold it. As of yesterday it is up +129% since the 8/15/09 signal and 25% since I took my profits. I have had other trades that I decided not to take for one reason or another (THD, EIS) that would have been wildly successful.
|
QUOTE (ben2k9) If we get a decent sized correction starting now, it will have been one of the most widely anticipated corrections of all time.
Really? The firm that I have a sizable chunk of my recent inheritance with has had me in all cash since April while they wait until "they can make sense of the market". They have recently (late Aug) been pushing me to allow them to start dollar-cost-averaging in, I have been expecting a correction and told them to hold off. I am also thinking of taking my business to another firm anyway for a variety of reasons beside this boneheaded missed opportunity, being in cash will make this easier.
I was directed to them by the person who handled my mother's financial affairs, it was not my selection although I went along with it and hoped for the best, but I am depressed and disgusted with their performance to date and told them so, in those exact terms. This has not endeared us to each other.
If I was as good as some of the regulars here (no comments from BB pleez) I would just invest it myself, but I am a suck@ss investor and trader. Trading large amts instead of my piddly low 5-figure Roth acct would only hurt my already poor performance.
|
|