TradersStudio Thread

Quote from TraderSystem:

Am wondering if we could make short notes of whatever we learned in your new user group.
More...

Of course you can ! Not only you can but you are absolutely welcome to do it !!! :D

Consider the Yahoo group as yours and post whatever you desire.

( ... and everything will remain stored in the archive for any future need ! There is an easy search engine to look for the desired topic )
 
Quote from Angelo_60:

Hi Murray,
As a purchaser of TraderStudio, I'm really interested in the algorithm implemented to figure out if high or low of a (daily) bar happened first.

Could you please elaborate on this subject or simply point out where to find additional information?

Thanks to everybody for this interesting thread.
More...



Up.

I just would like to know if this is inside TraderStudio (maybe a new version...) or it is proprietary or ... the reasion why this is an inappropriate question.


Not too much trouble, isn't it?
 
Angelo, I suppose that algorithm is proprietaty. Trade Station does not reveal how its algorithm works either. Roughly speaking they should be built around the same concept ( with some variations, perhaps ): the proximity of open and close to the high and low. If, for example, the open is near the low and the close is near the high it is logical to assume that the market made the low before the high. This is a simple example but there are all the intermediate situations which can be really more uncertain. I think both TradersStudio and Trade Station have run statistical studies about the occurrence of the open and close inside a daily bar and set up a special algorithm to determine if the low or the high occurred first based on these statistics. Since these statistics costed time, and perhaps also money, the algorithm is kept secret.
 
Quote from samuel_weber_74:

Angelo, I suppose that algorithm is proprietaty. Trade Station does not reveal how its algorithm works either.
More...
Samuel,
first of all, thanks you your reply.
In my opinion, it is better to receive a reply like" this is none of your business" than no reply at all.

Just to clarify:
- the Tradestation assumptions (that are important to me) are nor proprietary, but clearly explained in every Omega manual, from the first TS 3.5 I began using about 20 years ago (see jpeg joint);

- for my money, hope they know what they are doing is not an option. Historical backtesting is anyway a pale rappresentation of what you are going to meet in real time trading, so -whenever is possible - don't accept hope as an option.

- as I am a real TradersStudio owner, and not some droll who comes here just to repeat what it is written in the manuals with a tone of "great news" (as happened in the past), I can reasonably say I have no understanding about how/when TradersStudio uses this option (it is not only a matter of how it is calculated, you see?).

PS Normally when I talk about this subject, 99% of the people start talking about the "infamous" buoncing tick options and how it can affect your testing reults.
They are 100% right ..... and 100% out of subject.
I use these assumption to time entries (BUY & SELL SHORT), while the buoncing ticks is all about exits.... and I've never used it (my stops are never so close to the entry price).
 

Attachments

Quote from Angelo_60:

Samuel,
first of all, thanks you your reply.
In my opinion, it is better to receive a reply like" this is none of your business" than no reply at all.

Just to clarify:
- the Tradestation assumptions (that are important to me) are nor proprietary, but clearly explained in every Omega manual, from the first TS 3.5 I began using about 20 years ago (see jpeg joint);

- for my money, hope they know what they are doing is not an option. Historical backtesting is anyway a pale rappresentation of what you are going to meet in real time trading, so -whenever is possible - don't accept hope as an option.

- as I am a real TradersStudio owner, and not some droll who comes here just to repeat what it is written in the manuals with a tone of "great news" (as happened in the past), I can reasonably say I have no understanding about how/when TradersStudio uses this option (it is not only a matter of how it is calculated, you see?).

PS Normally when I talk about this subject, 99% of the people start talking about the "infamous" buoncing tick options and how it can affect your testing reults.
They are 100% right ..... and 100% out of subject.
I use these assumption to time entries (BUY & SELL SHORT), while the buoncing ticks is all about exits.... and I've never used it (my stops are never so close to the entry price).
More...

Bouncing ticks are a big problem in backtesting. By default TradersStudio takes worst case for bouncing ticks which means you will end up with two losing trades and we will assume the market if it closed near the low went from Low to High and then back to Low.

If you use the InitSD = True switch then we have a more normal bouncing tick. In TradeStation your allowed to change the bouncing tick parameter which changes your results, how can this be valid ?. What we do when this same day flag is true is to use a classic bouncing tick when we close in the upper end or lower end of the bar. If we close in the middle of the bar, we use a algorithm I developed which uses the past 1-3 days of price action to make a guess which happened first.

Backtested results always taken as a overstatement, which is why in TradersStudio I alway take the approach to understand results when in doubt.
 
Quote from TraderSystem:

Where is Gyles? Am waiting for the rest of the experiment. So far, it was good. :)
More...

Greetings to you all for the “New Year 2009”!

Sincere apologies for the long delay as I was away due to urgent work at home. I shall continue the experiment by next week or so.

samuel_weber_74, welcome to the TradersStudio family. :)

Sincere thanks for new yahoo group; a very good idea. It will prove to be very helpful. :)
 
Quote from samuel_weber_74:

Of course you can ! Not only you can but you are absolutely welcome to do it !!! :D

Consider the Yahoo group as yours and post whatever you desire.

( ... and everything will remain stored in the archive for any future need ! There is an easy search engine to look for the desired topic )
More...

Thanks a lot, samuel_weber_74. :)
 
Quote from Murray Ruggiero:

Bouncing ticks are a big problem in backtesting. By default TradersStudio takes worst case for bouncing ticks which means you will end up with two losing trades and we will assume the market if it closed near the low went from Low to High and then back to Low.

If you use the InitSD = True switch then we have a more normal bouncing tick. In TradeStation your allowed to change the bouncing tick parameter which changes your results, how can this be valid ?. What we do when this same day flag is true is to use a classic bouncing tick when we close in the upper end or lower end of the bar. If we close in the middle of the bar, we use a algorithm I developed which uses the past 1-3 days of price action to make a guess which happened first.

Backtested results always taken as a overstatement, which is why in TradersStudio I alway take the approach to understand results when in doubt.
More...

Are there any other cases where you underestimate back tested performance in your analysis? Is your goal is to make back tested results more realistic?
 
Murray,
over the weekend I had an idea. Instead of making webinars, why do you not write a newsletter ?? Something containing both trading ideas ( similar to Formula Research ) and ( in depth ) "tutorial lessons" about the use of TradersStudio. Personally I would be glad to subscribe. TradersStudio looks great and I want to make the best possible use of it.

I really hope the documentation for the new version will be substantially improved and increased. Just a (friendly)reminder. :D

I look forward to see the examples you promised. Would it be possible to see the virtual system in action ?

Wishing all the best. Sincerely.
 
You must log in or register to reply here.
Back
Top