Have any of you managed to create successful bots used for trading FOREX? I have a buddy swearing he could make one and asking me to invest a bunch of money with him into this venture but I'm a bit skeptical. What sort of inputs would the bot need in order for it to be able to take good trading decisions?
You don't need to invest anything to create a bot, just knowledge of programming and an investing strategy with which to implement. I have my own bot which I created with someone, and it's a fairly simple piece of software. It's Excel based, and once the commands are inputted into it, it can trade autonomously throughout the day. However, automated trading isn't a magical solution, and I prefer to have a few set command sequences that I can activate when I see an opportunity, rather than give free rein to the bot itself.
I write automation software, so the coined and common use of "bots" {scaled down automation) are very easy in C++ or VB. What you should be worried about isn't the bot, but whether the person is legitimate or not. There's a danger in using automated software if it's not done right. It can keep doing a certain action if it isn't programmed to stop. Like investing all your money away into something that is sinking. A human operator knows better and knows when to pull out if they see it dropping, but the computer may not be able to do that in all cases if it isn't programmed to know or see the signs correctly. I don't trust other people's software as a rule of thumb, and only use my own. His program may be okay, but if I were you and you still wanted to do it, I would have him prove that it was functional and that there weren't severe risks first before investing anything with it or with him to be safe.
None of us could code to save our lives, so yes, we would need to invest into this project. I have no clue how much though. We haven't looked at the figures yet. @eddie that is all good advice, I will keep it in mind when deciding how to move forward. Thank you.
I would think that you can get away with very little investment to get something to trade with. Perhaps look in some forex forums. I do some work with automated trading systems so I have some experience with this subject.
The thing is, most of the bots out there aren't succesful in fulfilling their job. Which is making profit. Quite simply a bot like that would make it's owner... well you would have pretty much infinite money. There are many who claim they have such a bot, there are many automated trading services that want to invest your money into those bots. And as I said, I very much doubt any of them work that well. I'm sure there are bots that make a steady profit when the conditions in the market are right though. Now that being said, an automated bot being the source of pretty much hassle free and endless money, don't you think a lot of really really smart and powerful groups of people (including all the banks in the world) have tried it? What makes you think your friend can pull this off? Since none of you know how to code you could just buy a ready made bot from someone. Then all you need is to input the magical formula that it trades with.
I concur with this. My computers are running 24/7 looking for suitable experts adviser combinations, millions of candidates created weekly, and once in a while one is worth a further look. To prepare an adviser that will function properly, even for a couple months, is a huge task. To eliminate curve fitting and make a robust adviser is a painstaking effort that very few can succeed with. There are many costs for suitable testing software. There is a heck of a lot more to this than writing code.
For individual traders (like myself) a bot can be useful because the transaction value relative to the whole market is tiny. The problem for large trading houses is the issue of scale: what might work for me putting a few hundred through the system wont scale up when you're putting millions through it. Whereas I can trade the market structure and my trades don't impact that structure, a trader issuing huge orders will in itself change the market, and bots are notoriously fragile when this happens. Minor flash crashes are often attributed to this bot-malfunction phenomenon.
That is a good point. Whereas I view the ability to affect the markets as a bonus, I can see why it could cause issues when people are letting the robots do all the work.
Exactly. If you're really interested in the negative impact that bots can have on the markets, just read a little into the most famous "flash crash" of all which happened in 2010. https://en.wikipedia.org/wiki/2010_Flash_Crash#Early_theories The theory that a whole system of bots from multiple trading houses went rogue appears the most plausible to me. Obviously, since then someone has been arrested for fraudulent trading, but I still think the fact that automated trading was left unchecked ultimately caused the problem. Interesting stuff if you're a bit of a nerd.
