StockFetcher Forums · General Discussion · MOO orders<< >>Post Follow-up
ellios47
6 posts
msg #39473
Ignore ellios47
12/12/2005 12:10:42 AM

I've recently started using SF. I've got 2 questions of there is anyone out there who is knowledgable in the areas and cares to respond I would be very grateful.
First the statistics for the system that I've built has shown in backtesting 2 years to have a ROI of 343% with an average hold time of 1 day and 452 trades in 487 days. Is this a good return?
Second, The system requires that each stock picked by the system is bought at the open of the day following the trigger. Is this very realistic? What is the best way of going about buying nearest to the open? Waht type of order? Market at open orders?
Please excuse my lack of knowledge in this area, I'm still learning.
Thanks.
Eli


chs245
2 posts
msg #39476
Ignore chs245
12/12/2005 3:48:03 AM

I'm also a new user of SF and curious to know about other's ROI's. My best system has an ROI of 680% for 500 trades with a holding period of 1 day. Risk to reward is 2.06
Trading the open depends on the average daily trading volume olf the stock and of how many shares you want to trade. For liquid NYSE/NASDAQ stocks, an MOO is a very good option. For less liquid stocks, I would still trade an MOO on NYSE, but for Nasdaq I would be careful. Better to work with limit orders in the market. On the NYse, the specialist will provide you with the required liquidity, but you have to pay up for it.
Hopefully other users also will post their trade statistics. By the way, ROI may not be the right profitability measure. My average gain per trade increase as my holding period increases, but my ROI goes down, because ROI is calculated on an annual basis.

Oliver



BFreshour
139 posts
msg #39478
Ignore BFreshour
12/12/2005 9:20:40 AM

Most backtests that have a high ROI are not really tradable. They don't factor in commission costs or slippage which can be devastating to your profits.


ellios47
6 posts
msg #39486
Ignore ellios47
12/13/2005 12:44:04 AM

Thanks for the input guys.
Oliver it sounds like there is some potential with your system. The backtesting feature is interesting, however I am convinced of its credibility. I'll follow it in real time and then decide if it is a realistic trading system. I am not concerned too much with the commission costs because my system averages less than 2 trades per day. Oliver, your system also generates relatively few trades with great returns as well. I am concerned with the slippage. It would be great if there was a feature that reflected how realistic you rystem is. Is there anyone out ther that has an idea on how to determine ifyour system can realitically be implemented?
Thanks.
Eli


ellios47
6 posts
msg #39487
Ignore ellios47
12/13/2005 12:47:15 AM

As for the liquidity of the stocks that my system generates I use a filter that requires that the average 3 mo volume multiplied by the last price is greater than $1,000,000. Would you consider stocks with this criteria to be liquid? What would you consider to be a benchmark for large liquidity.
Eli


chs245
2 posts
msg #39489
Ignore chs245
12/13/2005 2:50:14 AM

Eli, $1m average dollar volume is very low. $10m is much more realistic. but again it depends whether you trade shares in multiple of hundreds or thousands.




BigOh
23 posts
msg #39628
Ignore BigOh
12/17/2005 11:23:21 PM

As for buying at the open. There are several things to be considered when using a MOO order. Order imbalances, news on stock and sector, opening indication and how the stock opens. For NASDAQ, I would wait till the bell then just hit the offer. NYSE and AMEX require a little more thought, specialist or as I refer to them, thugs, usually use the opening print to help their own positions or entry/exit of positions. At the same time the "Street money" also uses this time to get the larger part of their positions on. I think a case by case approach would work best in these markets. As for the OTCCB, have no clue but probably most like NASDAQ


judgetrade
106 posts
msg #39730
Ignore judgetrade
12/22/2005 7:27:21 AM

You have to be very carefull using the SF backtesting mashine, not only because of commision, bid ask spread and volume (with a litle experience you can judge that part!). The problem is, that it gives you wrong backtesting results:

Take the following screen and backtest ist (3 days holding period)

Show stocks where RSI(2) is below 1
and close is between 0.1 and 1
and roc (1) < -5%
and average price (1) * average volume (1) > 200000
and average price (100) * average volume (100) > 100000
close below open

For example CLYW:

Backtesting shows, that the trade signal was on 11/22/2005 @ 0.39.

If you test it by hand (homepage of SF -- > Create --> paste the code into "Customize your filter" and run it with and offset of 22--> It shows the trade CLYW at a price of 0.54 22 Days (or better Filter results for Friday November 18, 2005) ago.
Then on offset 21 (November 21, 2005), then offset 20 (November 22, 2005).
On all three dates the stock apears as a pick!

But the backtesting mashine waits (in hindsight!!!!!!!!!!!!) 3 more days and buys the stock at 0.39 at the open (0.42 at the close one day ago in the evening), e.g. at November 22, 2005.

I put that to the attention of SF in the Bug Forum. So far SF could not have cared less! (Maybe its my bad english!)

The right application would have been three trades
at

November 18, 2005
November 21, 2005
November 22, 2005

I still test everything by hand because of that (monster) flaw.
I go back 1000 Days and say the performance measures loud, so my girl friend writes them down into an excel sheet. And we can have a look at the equity curve afterwards.

Its a dangerous backtesting mashine!


john0024
7 posts
msg #39731
Ignore john0024
12/22/2005 12:22:32 PM

I've recently joined SF and have found the tools useful in the respect to providing me with monitoring signals. Let me explain. I use a RSI / MACD crossover match as my personal trigger to buy/sell. I use the SF to find stocks that are currently in RSI and/or MACD positions that might bring a signal in the future. IE: RSI <20 and/or MACD <-.2 could be inputted as part of a search on SF. I take the list/symbols generated and then track the stocks until my personal trigger formulas are met, then I buy/sell accordingly.

I too have noted that if I run a backtest using the SF 'Backtesting" mechanism I do get different results than if I paste the code into the 'filter builder' and use the "and offset of X" to test forward from the same date. (X being the days you want to test forward from)

Sure there are other products out there that'll do much of what the SF site does but you will pay some pretty hefty sums for many of those. My advice is not to trade the signals but use the symbols generated to bounce off your own personal trigger parameter.

john


StockFetcher Forums · General Discussion · MOO orders<< >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus