RE: [EquisMetaStock Group] Low Volatility Stocks...RSI, MetaStock Email Archive

Hi Steve,

Looks like some excellent results! I have a very similar plan on the S&P500 using long term Moving Averages with vertical % adjustments as upper and lower bands which seems to work very well.

You got me intrigued to look into your plan. I looked into Standard Error bands and found a complex formula that still has some minor errors in some calculations. ..

http://trader. online.pl/ MSZ/e-w-Standard _Error_Bands. html

I was able to modify these with optimisations but quite frankly I don't trust the original formula and it's too complex for me to invest time into resolving.

Can you offer any site reference or simple formulas to calculate standard error bands so I can do some backtesting please?

I think it would work very well on Gold on an hourly chart which is my forte. Gold has some similar robust characteristics.

Do you trade any other systems or do you only use the one and find stocks to fit it?

Lee.

To: equismetastock@ yahoogroups. com
From: skeeter47@xxxxxx com
Date: Fri, 27 Jun 2008 10:02:47 -0700
Subject: RE: [EquisMetaStock Group] Low Volatility Stocks...RSI

Lee,

Analyzing divergence is an art form. Over thirty years, I've learned that I'm not that great of an "artist". Everything that I have designed is mechanical. I've been accused of being a "mechanical monkey"...a title that doesn't bother me at all. The trick for most traders to take the emotion out of trading. Nothing takes care emotion like a set of rules to follow. That doesn't mean that traders will follow their mechical approach and try to outsmart their own rules. That's akin to trying to figure out divergence (subjectivity kills most speculators) .

Regarding the chart that I sent: The bulk of my trading is in the futures market. If you have an approach that works, you maximize your returns by moving to the markets that afford the greatest leverage. I'm not a crusader for futures, but it works for me and commodities have always been my love. When I was a broker, 95% of my biz was in the futures market. As a CTA, I traded only in the futures market. So, I feel very comfortable with the double-edged sword of leverage.

I trade eight or ten different markets with a very robust approach. All entry and exit strategies are the same for all of the markets I trade. What a concept: using the same approach on grains, interest rates, etc. The only difference in the rules are slightly different trigger levels on entry and exit levels (I think due to volatility and the nature of the individual markets..."cocoa isn't the same as the 10-year note"). Stops vary from $300 to $800...also, depending on the nature of the volatility of the individual market.

For years, I was a swing trading momentum guy. I still believe the StoRSI that I apply to the markets is the best tool that I have ever developed/borrowed/ stolen. Believe me, I have tried and tested just about everything. I have over 20,000 hours of testing, in MetaStock, applied to momentum oscillators. If there was an oscillator that was more consistent, I would be using it.

The problem with swing trading with momentum oscillators is that once you are in a winning trade, the momentum triggers exits you prematurely ... before the big money is made. This sets up circumstances that I have struggled with over the years. I have always been able to produce a high win/loss percentage, but have always had average win to average loss rations just above 1.00, When markets are trending, you take a small profit and maybe (depending on your rules) you reverse your position (contratrend) and immediately get stopped out. This causes very nice profits in choppy markets and can get ugly in trending markets.

Fund managers have always taken a different approach. They play break-out and various trend following methods and produce a low winning percentage (many times in the low 40%'s or lower), yet have a very high average win to average loss ratio. After slamming many numbers into sharpe ratios and ulcer indices, I finally (duh) came to the knowledge that I must change my basic though process and trading approach.

At the beginning of the year, I switched to a momentum oscillator entry and a trend following exit strategy. This strategy produces 40-50% winners, but the average win to average loss ratios vary from 3 to 9. The key to my current approach is diversification and the sharpe ratios it develops.

The blue line on the chart that was send was the moving average, RSI oscillator that Super forwarded. I've attached the long bond chart, without the blue line, and will walk through each signal:

1. The overview is quite simple. I initiate a position the opening after a contract penetrates a trigger level on the StoRSI (on 1/3, I shorted bonds...due to the close above the trigger level on the previous day). Unfortunately, the position was stopped out the following day for a $600 loss.
2. Everything is "reset" and I wait for another StoRSI penetration of the trigger levels. This occurs on 1/22 and on 1/23 I entered a short position.
3. After I am in a position, I completely ignore the StoRSI and I wait for a penetration of the opposite SEB (Standard Error Band). In other words, if I am short, I am looking for a penetration of the lower standard error band...if I am long, I am looking for a penetration of the upper standard error band). Nothing will take me out of the position unless it is stopped out or the market causes a violation of the bands.
4. Once I get a close above/below the bands (depending on which way I am positioned), I exit and move to the sidelines. The exception: if a close is above/below the trigger level AND on the same trading session the StoRSI also penetrates it's triggers, I reverse the position. On 2/19, the market causes a close below the lower SEB AND the StoRSI penetrated its lower trigger level. These combined circumstances caused a reversal of positions on 2/20. The same circumstances apply to the next trade: on 3/20 the market closes above the SEB upper band (the signal to close the long) AND the StoRSI punches throug the upper trigger...causing a reveral of positions and a new short position.
5. Although the StoRSI penetrates its lower trigger many times during the nex two months plus...I am only going to exit if the close is below the lower SEB. This takes place on the 6/12 and I exit and move to the sidelines. I don't reverse, because the StoRSI hasn't gone below the trigger level.
6. It doesn't take long for the StoRSI to drop. The next day 6/13, the StoRSI closes below its trigger and I jump back in a long position after being on the sidelines for only a day.
7.This morning, 6/27, I covered the long position (yesterday, the close was over the upper SEB band). I am NOT going short. The reason is in the rules: I need a penetration of the trigger level to initiate a short position. Even though the StoRSI is above the upper trigger level, I am not interested in a new trade unless the previous day's StoRSI had traveled from neutral territory and has moved through a trigger level. So, in this case, to initiate a new short position, the StoRSI must move below the trigger level and move above it once again. At that point, I will put on another short. If the market continues down and drags the StoRSI below the lower trigger level, then my next bond trade will be a long position.

I know that this windy pontification is probably not the best explanation available. But hey, I was educated in the Dee-troit public school system and they taught ebonics. Sorry, if it is not crystal clear. Hopefully, it sheds some light on how to combine momentum oscillators and turn the position into a trend following system.

Most of the SEB settings hover around 55 periods and the triggers for the StoRSI are optimized to 5's (i.e., 20-80; 15-85, etc.) All are symmetrical. Asymmetry will test a lot better, but only a fool would use asymmetrial triggers.

I am trading 5, 10, and 30 year rates; corn, wheat and soy meal; cocoa, coffee and sugar. The approach works with just about all the futures contracts ... including the mini's. But, when you examine the sharpe ratios and the drawdowns, you must draw a line in the sand and chose the best performers.

Hope this helps,

Steve

--- On Thu, 6/26/08, Lee Lucas <leeontherun@ hotmail.com> wrote:

From: Lee Lucas <leeontherun@ hotmail.com>
Subject: RE: [EquisMetaStock Group] Low Volatility Stocks...RSI
To: equismetastock@ yahoogroups. com
Date: Thursday, June 26, 2008, 10:18 PM

Hey Steve,

I have not been able to successfully backtest Divergence. I think it more for chartists than formula following backtesters like us.

I can't tell anything from you picture other than the fact that your system is working brilliantly. Can you tell me what kind of indications your used to open and close on this system?

An MA of the RSI should follow the RSI tightly and basically remove all of the minor bumps. It helps remove double and triple signals when the RSI jumps up and down in the overbought and oversold areas. Perhaps use a 7 or 14 day MA of the RSI. The blue line looks like it has a 180 day MA applied to the RSI.

The whole idea of this methodology is to remove a few losers. I also use a formula to say that the MA of the RSI must also be pointing up before taking the long.

Lee.

To: equismetastock@ yahoogroups. com
From: skeeter47@xxxxxx com
Date: Thu, 26 Jun 2008 08:59:04 -0700
Subject: RE: [EquisMetaStock Group] Low Volatility Stocks...RSI

Lee,

The blue line, in the upper frame on the chart, is what I think is your RSI....is this similar or matching to your formula? Keep in mind, divergence is a very nasty trap and can only be applied randomly as it occurs. Divergence equals subjectivity. How much does an issue diverge before you can safely time yourself into a trade (a little divergence, a little more divergence, or ad nausem divergence)? Many educators teach divergence with certain momentum oscillators (RSI, CCI, etc.). But, none of them ever define how much is the right amount of divergence. I'll stick with a faster, more reliable and profitable StoRSI.

Your thoughts?

Steve

--- On Wed, 6/25/08, Lee Lucas <leeontherun@ hotmail.com> wrote:

From: Lee Lucas <leeontherun@ hotmail.com>
Subject: RE: [EquisMetaStock Group] Low Volatility Stocks...RSI
To: equismetastock@ yahoogroups. com
Date: Wednesday, June 25, 2008, 5:18 PM

Hey Preston,

The thing I like about RSI compared to most other indicators is that other indicators tend to be lagging indicators where as an RSI can show a slowing or change in direction earlier than the price shows.

In combination I have also made a moving average of an RSI i.e. Mov(RSI(C,14) ,40,E) and looked for cross overs. This can cut out a lot of noise. Works similar to a MACD I suppose.

Further more Divergence which has been quoted as being the most robust and secure oportunity to take a trade uses RSI or MACD. For those who don't know of Divergence - this is when price continues in it's direction while the other indicators (RSI or MACD) have already turned to go back. The assumption is that price will then turn back to keep up with the indicator.

Lee.

To: equismetastock@ yahoogroups. com
From: no_reply@xxxxxxxxxx s.com
Date: Wed, 25 Jun 2008 14:27:21 +0000
Subject: Re: [EquisMetaStock Group] Low Volatility Stocks...RSI

Lee,

I would love to hear more about how you are using the RSI and ATR.

Also wondering if you have read Baeyens book on the RSI?

Preston

--- In equismetastock@ yahoogroups. com, Lee Lucas <leeontherun@ ...>
wrote:
>
>
> It may be much easier to understand and see on a chart if you can
ship the backtest formula.
>
> I play alot with RSI and ATR
>
> Lee.
>
>
> To: equismetastock@ ...: andysmith_999@ ...: Tue, 24 Jun 2008
03:55:13 +0000Subject: [EquisMetaStock Group] Low Volatility Stocks
>
>
>
>
> I've been experimenting with this methodology: Every weekend I use a
screen to get stocks with certain fundamentals( based on P/S, P/E,
etc). Then I rank the list by 13 and 26 weekrelative strength and
discard any stocks that are not near the top onboth lists. This
leaves me around 20 stocks per week. Then, I look at the 20 charts to
see which charts have nice, smooth,up trends. I each chart is
displayed:-- ATR(10), this is used to calculate position size-- ATR
(10)/Close, this is used to get a sense of volatility-- ATR(10)/ATR
(50), this is used to get a sense of "is the stock morevolatile than
usual"The ATR(10)/Close usually works out to be around 3.5% to 5.5%.
Here'sthe part I'm still grappling with: I ignore any stocks above
4.5% sothat I end up with the smoother, tighter up trends. Please
share anythoughts on this.PS. I owe much of this to Roy's newsletter
and Super's posts over thelast few years.
>
>

at CarPoint.com. au It's simple! Sell your car for just $30

Find out: SEEK Salary Centre Are you paid what you're worth?

Check our comprehensive Salary Centre Overpaid or Underpaid?