Murray Ruggiero
Sponsor
TradersStudio can do that with addin products we sell
NeuralStudio, Genetic Optimizer
I'm at step 0 regarding systems.
- Thread starter rainmonkey
- Start date
TradersStudio can do that with addin products we sell
NeuralStudio, Genetic Optimizer
It may have already been mentioned but look at
Tradestation.com
Easy Language. Everything you mentioned can be done there.
Tradestation.com
Easy Language. Everything you mentioned can be done there.
They will all come out and say they can do it...tell them to show you.
ONLY with the addition of add-in products can you do most of these things. You cannot mine your results. You cannot back-test multiple symbols simultaneously. The entire reason I developed my own was because TS couldn't do it.
Your results aren't stored in a database and the time it takes to back-test is unbelievable.
ONLY with the addition of add-in products can you do most of these things. You cannot mine your results. You cannot back-test multiple symbols simultaneously. The entire reason I developed my own was because TS couldn't do it.
Your results aren't stored in a database and the time it takes to back-test is unbelievable.
@Tr
your own using TS with your own created Program to go with it or are you using a platform you created for yourself that does everything your looking for and rivals TS.
Either way thats cool but I just wanted more clearity so I can add it to my trading IQ of possibilities
your own using TS with your own created Program to go with it or are you using a platform you created for yourself that does everything your looking for and rivals TS.
Either way thats cool but I just wanted more clearity so I can add it to my trading IQ of possibilities
While many professionals do use TS, it is a platform that is aimed for the mainstream. During the process, they left out many features. Many of these were added by 3rd party vendors. TS is extremely slow in back-testing.
Nothing that I have mentioned was available 3 years ago in TS that was either easy to implement or fast. I attempted to code my strategies in TS and while they will tell you it can work with multiple time frames, they obviously didn't test it very well. At the time, I knew the patterns I was looking for and TS did not locate them. I know this because I had used spreadsheets and other tools to tell me when the patterns occurred. When I went through TS back-tests and did not see the results for days when I knew a pattern existed, I decided it was not reliable enough for me.
While I was very proficient at easy language, I hired a TS professional to code a stop-loss that I wanted to implement and he said that it could not be done. Maybe he lied to me and he's been using it all along. This guy did program for a very well-known trader, so if he found that it was useful maybe he just wanted to throw me off.
There are TWO reasons why I didn't trust TS. Throw in all the data-mining, slow back-testing, and I know it's not for me.
Instead of giving up, I decided to start from scratch and do it on my own. I designed the UI and sent it to my programmers to connect it to the api. I developed it to be easy, fast, and self-reliant on all aspects and there would be no need for a 3rd party product to make it better. I spent three years developing it, but I initially had used a product that traded only, that I had used shortly after leaving a hedge funds prop trading group. That program was trashed and I started at ground zero.
My goal is to expose to people here that scalping is not the only method of day trading. I also want to show people how to mine data, and use statistics in their trading. In my example I will really be testing the data to see if the Pivot Points are relevant and is there a need to focus on them. I have already shown that the open is relevant as well as the high / low / close in the case of gaps.
I am currently not interested in selling anything to anyone as I am more interested in if someone uses my product, that they are successful. That's how I differ from everyone else...
@TR
Thanks!
Thanks!
This is a serious accusation against Tradestation Technologies. You may be right given the history of the products, for example TS2000 had a defective backtester but I donât know, you may be also wrong. But this is too serious. Are you telling us that TS fails to identify patterns people are looking for and they are there using other methods of identification? This is a disaster⦠Please provide specific examples and I hope the TS people will come out and try to clarify this.
Now, some other things you are saying do not make sense. There are enough products out there for traders willing to climb the learning curves and use them. Serious traders do not spend their time developing platforms from ground zero. It sounds you want to become a vendor. What you said about data mining and its importance is correct but there are few products out there you can use with TS and other platforms that offer an integrated framework for data mining/backtesting/system development/execution. No need to reinvent the wheel IMO.
I have given you my experience with TS. Is this everyones? I doubt it. I was attempting to do something very different and at the time, combining time frames was not working for me. I spent hours going through the back-tests, sent the code to someone to see if it could possibly be an error, and still no luck. There was no answer to my problem, so instead of giving up ... I was on a quest to answer my questions.
I no longer have the code as I haven't used the platform in years. If I find it, I will be certain to send it to you.
It doesn't make sense to who? That I want some questions answered and they can't be answered with a readily available product? Nor can I attempt to mine data as I so desire?
I haven't re-invented the wheel, I have created what I was looking for because it did not exist in a single platform. My product answers the questions that I have had. Therefore, my time and effort are totally justified.
I am certain that there are products out there that do things better and for certain are prettier, I am only interested in the aspects that I mentioned.
Are you implementing day trading strategies or larger time frames?
I totally agree with multiple systems and switches. If the market opens in different places, you may want to trade differently, if your back-testing has proven to be more profitable. With the database, your program can find out what the best method to trade is and trade accordingly. This could be done in seconds using a query to the database along with a PHP script. While I currently don't implement this, it is certainly on the to-do list.
I also developed my platform to allow me to enter a stock as both a long and a short. This way, I have no bias and the program trades based on the price action. I will say, this was a very difficult portion to implement and troubleshoot. This is mainly due to all of the variables with re-entry after a stop-out.
The platform isn't just based on my method. The platform includes:
15 Different price action based entry methods (with multiple variables that are all adjustable).
3 Different methods for re-entry after a stop-out and this can be turned on or off
9 Different Profit Target Methods which is set as a % of the position and can be set to 0% if you don't want to use a Profit Target
10 Different Stop-Loss Methods
When back-testing and in live trading, you can enter a specific number of shares or a $ amount of risk for determining shares to be traded. What this will do is allow you to risk a specific $ amount / trade and the number of shares traded will be calculated automatically by the program based on your stop-loss. Obviously, the stop-loss amount must be a calculated move.
All of these are totally adjustable and you select each one independently and you do this in the user interface. The combinations are virtually limitless. Not to mention, eventually a trader will be allowed to enter their own price-based strategies in C#.
What I have learned is that the key to success with most of the methods is consistency. Don't go chasing what worked today, because it may not work tomorrow. The market doesn't move the same way everyday, but I know that the probabilities of it moving in my direction for at least a certain $ amount are pretty good. The key is what's it going to do the rest of the day and while no one knows, the probability of certain patterns forming on the daily chart are just not likely. This is what you find out when you devise a data-mining application that looks for patterns on the daily chart. While I am still developing this, this is what I have learned so far...
What % of the time should you expect a continuation after a trend day?
Does a particular pattern often precede a trend day?
Are all the patterns that most people are looking at even relevant?
What does the open tell you? From my research, the open is very important.
The platform is really intended for day-trading only and this is another reason it was developed from scratch.
You sound like you are selling