[nave]

Originally Posted by BlowFish »

What tools have you been using to design your strategy? Excel or something? How complex is the logic for the system?

I'm totally new to this terminology, so can you please elaborate what you mean by tools? (sorry for my ignorance) and regarding the logic, I don't know how to grade...it is complex thats all I can say...

Originally Posted by BlowFish »

There are solutions that can be 'scripted' in a fairly straightforward sort of way.

Thats a good news. easy solution is always acceptable

Thanks for your contribution

[BlowFish]

By 'tools' I mean what did you use to design your strategy? You say you have been "working on a strategy for a log time". I wonder what you have been doing and what you have been using to do it and, without giving the game away a general idea of the result. (the type of system).

If your systems is complex you probably won't need to do much in the way in optimisation. Optimising complex systems needs great care to avoid curve fitting. Curve fitting gives fantastic results on the data used for the optimisation and abject failure on out of sample data.

What your system is will dictate the best tools (bits of software) to use to automate it. Does it use traditional technical analysis? or maybe simple price action? Does it look at correlated instruments or maybe it is based on statistics or some sort of qualitative approach. Maybe it requires fundamental information as an input too?

I know you asked for answers that where not "it depends on your requirements" but for anyone to give anything but general information they would really need to know more about your strategy and exactly what you want to do. In any case that is the correct answer (it depends on your requirements).

The other big question is whether you want to go unattended automation or you are going to keep an eye on things?

Forget all the languages platforms optimisers packages and what have you. Approach it from the other end set out what you are trying to achieve the more specific the better.

Oh one thing I notice you asked and has not been answered what else to things like tradestation, Multicharts etc. offer and the answer is they are charting packages. Ninja started off as an order entry software but has developed into pretty full featured charting package. They all farily general purpose and have quite a bit of overlap but there are odd things one or other does better. Based on your requirements all might be equally OK or it might steer you in a particular direction. It might rule all out in favour of a more specific package or a more bespoke approach.

[pradeep]

Blowfish,
Sorry for the delay in reply. I did not use much of software... simply a charting software to observe the price action, volumes and other data... Till now I traded on a single instrument. I designed my own indicators. I used excel to apply these indicators on data. All the calculations runs in my mind. But the problem here is, I have to spend every minute observing the market clearly. This prevents me from looking on to other instruments and maintaining a portifolio. That's why I came for automation. Though the strategy is working fine currently, I came to know that optimising the strategy would give better results...Yes Im aware of overfitting problems and I want to take serious care of it.
However, I dont want to be at the mercy of my computer...I'll surely have a look on whats going on.
Thankyou soomuch for your advise...Im trying to work from the other end right now... previously Im dont know where to start and what to start...but now im a bit aware of whats going on and what to do...

[nave]

Blowfish,
sorry for the delayed reply. I did not use much of software to design my strategy. I simply used a charting software and a data feed for the market data like price change, volumes etc...
I framed my own indicators by generally avilable market information. During implementation, all the calculations run through my mind. however, I used excel for little support during my implementation. My strategy is working fine giving me profits.
But the cons of it are, I can only apply it on a single instrument as it is required consious observation of market. Thats why Im trying to automate it so that I can look in to other instruments. However, I dont want to be at the mercy of my computer and I would be frequently supervising the system.
Further I used certain parameters whose value is set by me based on my experience with the instrument. But while applying the strategy to other instruments, I have no enough experinece with them to set a value for these parameters. Thus optimisation is required.
Yes, Im aware of the overefitting problem. I'm sure to take care of it during optimisation process...thanks for alerting me...
approaching from the other end...thats a valuble suggestion...thankyou soo much... Im trying to do the same...previouly I dont know where to start and what to start...but now I understand whats going on and what to do...

[nave]

THE EVALUATION AND OPTIMISATION OF TRADING STRATEGIES by Robert Pardo

People new to strategy building and optimisation process checkout this book...it can be useful in giving a basic idea about these concepts...