You thoughts??

[Doji3333]

We're obviously in a correction of the major trend (bearish). The markets have stalled to almost "even" for the last 60 days.

I've heard a lot of trading systems have lost BIG in the last 3 months.

How are your system's doing?? Do you have any systems?

I'm curious because I'm running robots in efs and want to know how people are doing thru these tough times??

If you want to share, it would be appreciated??

[Doji3333]

Well, I guess most people don't want to discuss their systems because there have been no replies to this thread.. So I'll start by discussing one of my systems...

I have a number of robots running on my PCs. My latest robot is called the "scalper". I wanted to build something with very high probability and decent returns. After all the years and many MANY attempts an building profitable trading systems, this took me about 5 months to create (and is still not perfect).

The concept is to find and execute trades while high probability trends are happening. Of course, this is not 100% accurate all the time. The other method I'm using is what I call "confirmation bars". This is a method of identifying if the trend is continuing as projected.

Right now, I'm using only the basic features of my code to find the trade triggers. I also like running my code on 10 minute charts. I find 10 minute charts provide good trading opportunities and provide limited risk overall.

Here is a screen shot of my robots running in esignal. Running it with the default settings, it averages about 70~85% accuracy while trading. There are many different ways to set up these robots, but I like to run them in a very conservative mode of operation.

If any of you have specific questions about building advanced trading systems into EFS, I'll be happy to try to answer your questions (without divulging any of my specialized techniques.... I'm sure you understand why)..

[spyder9000]

Hey whats up Doji,

Do you have any systems or ideas that are designed to properly use Bollinger Bands?
I trade the classic (perhaps incorrect) way...which is short if the stock touches the upper band and get long when it tags the lower band.

Combining this with stochastics, my success ratio is embarrassingly low, and I've been losing money.

Can't seem to find any ideas on these forums regarding the BB, hoping you can offer some advice.

By the way I just registered for Esignal a month ago, so I'm a bit of a rookie.

[Doji3333]

I've used alot of Bollinger Band stuff in my code and other client's code. It seems to be a pretty common component of trading systems.

Basically, the outter BB levels EXPAND as the market increases trend strength and contract when the markets are narrow/flat. Given this information, you would be looking to execute trades when congestion/contraction of the BB outter levels ends - and begins expanding again. This is the "timing" feature of BB at work.

There are other ways to use it, but this is what I consider a stable B timing mechanism. Now, the middle BB level is a MA - so you can use a test of close() > or < the middle BB level as a base trend filter.

Including the Stochastics into the equation would have to be done in base terms first with simple rules...

S_K or S_D > 50 = bullish
S_K or S_D < 50 = bearish

then

(S_K > S_D) within last 2~3 bars = bullish
(S_K < S_D) within last 2~3 bars = bearish

Now, you have to decide how you want to develop your system.. Find the highest probability triggers and then identify the max/min/avg/best trade variables...

For example, you might find that getting in and out with 1~3 pts profit is best. You might find that only buying when price is above the BB_MA and a crossing of a STOCH (set relatively short term) will generate good additional (trending Bull/Bear) triggers.

Given these components, you'll then have to decide what you want your robot to do and when.

Generally, when the price is above/below the BB Middle level, then consider the trend bullish or bearish. If the BB outter levels are expanding after a congested range, then you could be seeing a new trend establish - so you want to try to trade the price expansion - right??

Use the STOCK hooks to try to catch the best triggers in a trend and you should have a basic system in place.. Try to buy Sto X UP below 50 or 30 when the price trend is up and try to sell Sto X DN above 50 or 70 when price trend is down.

Using a simply BREAK EVEN stop will help reduce losses. Try BE + 1 or 2 ticks to help keep your losses in control.

[Doji3333]

Hey, I've got an idea. Let's explore your idea here on this forum. I'm pretty sure it's OK with esignal if we post code/charts as we consider your thoughts??

Let's start by marking up this chart with out thoughts and go from there...

[Doji3333]

Here - check this out.. Simple logic to define a trend analysis model...

When MACD, Stoch 1 ~ 2 are all hooked up (or down), define a trend direction (colored areas).

Now, the trick is to use the MAs and BB Mid level for stops and the short term Sto (going against your trade) as a stop activation mechanism.. The only trick with this is where you want to adjust your stops to...??

This is just a start...

B

[spyder9000]

Whats up Doji,

Actually I was hoping for some advice or direction on how to properly use BB. It looks like your charts and suggestions are designed for advanced traders, so I'm not sure how to interpret it.

For example, I been shorting every tag of the upper bands, but sometimes the stock keeps "riding" the band up.

I can't seem to find any advice or trading strategies on the most popular ways people use the BB, so I was hoping that you could advise, or direct me to some stuff on either your website, or someone else's.
Thanks.

[spyder9000]

By the way....do you know where I can find a Esignal formula for alerts on when a stock tags the upper or lower Bollinger Band? Having some trouble finding it in the forum, and not sure who or how to ask.
Thanks.

[waynecd]

Hi,

I hope my chiming in is ok.

When MACD, Stoch 1 ~ 2 are all hooked up (or down), define a trend direction (colored areas).

Consider adding divergence of the 14,7,3 stochastics %K only as a supporting or confirmatory signal.

I'm in the process of trying to duplicate your background coloring efs so we can banter about this approach, that is, if you're still interested in doing so.

wayne

[waynecd]

Hi spyder9000,

I found this via forum search:

see attached

wayne

[spyder9000]

Whats up Wayne.

1.
Yup, we definitely should discuss the BB theory further...your thoughts and suggestions are more than welcome, along with anyone else who'd like to share.

2.
I appreciate that link regarding BB alerts...where'd you find it?
That particular study looks like it offers a lot more alerts than just tags of the upper and lower bands, but I'd be interested in checking out other BB alerts.

3.
Do you use BB in your trading? I've found stochastics to be a very reliable indicator for counter trend trading with BB, but I've never used divergence as a confirmation.
How do you implement it, and do you use any other indicators?
Thanks.

[waynecd]

Hi spyder9000,

I placed the link to the originating forum post as the first line in the bollingerbandsalert.efs study.

The study only alerts when price is above the upper BB or below the lower BB using a percentage calculation.

I appreciate that link regarding BB alerts...where'd you find it?
That particular study looks like it offers a lot more alerts than just tags of the upper and lower bands, but I'd be interested in checking out other BB alerts.

I don't use BB so any script additions, tests, etc I would need to code. I do use stochastics. I found divergence is not very consistent just helpful in pulling the trigger if it confirms. For BB and using stochastics to show possible reversal pivots consider using a stochastics/price divergence indicator. Attached is a simple indicator that plots divergence for RSI, CCI, & Slow Stochastics.

Do you use BB in your trading? I've found stochastics to be a very reliable indicator for counter trend trading with BB, but I've never used divergence as a confirmation.
How do you implement it, and do you use any other indicators?

[Doji3333]

yes, still interested in doing this. Been working on other projects too. I had a bit of a family emergency about two weeks ago. My Mother had a massive stroke and died. We were all shocked.

Please forgive me as I have been trying to catch up with all my work and deal with the obvious. Let's see what we can do with this??

Originally posted by waynecd
Hi,

I hope my chiming in is ok.



Consider adding divergence of the 14,7,3 stochastics %K only as a supporting or confirmatory signal.

I'm in the process of trying to duplicate your background coloring efs so we can banter about this approach, that is, if you're still interested in doing so.

wayne

[waynecd]

Hi Doji3333,

I'm sorry for your loss, please accept my condolence.
I'll keep an eye on this thread. Take your time.

wayne

[spyder9000]

Doji,
That is terrible news, I'm very sorry to hear that.
Please extend my condolences to you and your family.