Recent content by windsurfing_stew

This was my post. I've just realised my figures are out of date. Costs and revenues are about 2x to 3x larger now. Still, the thought remains the same.

Currently TWTR has about 200M active accounts. It's market cap is $35,000M. This means that each active account is valued at $175.00. TWTR's current revenue is about $300M, ie about $1.50 per active account and their expenses are about -$500M, ie about -$2.50 per active account Now clearly...

I use GFT (www.gftuk.com)'s API for automated spread betting. There is a Java version and a COM version. FXCM also have a spread betting API but their range of products is more limited.

How did you get on with your investigation into Neural Networks?

Also worth noting that running the same test on two different machines may produce different solutions because - 1) Often computationally expensive operations are optimised for different chipsets - eg AMD vs Intel. 2) The seed may be different each run.

Make sure you keep the start and end dates the same too.

Using Neuroshell Trader in the most part.

Perhaps add more FX pairs to your strategy (say another 9). Keep the best few pairs from optimisation for your out-of-sample testing. Usually if a model works on one pair then it won't work on all pairs but will work on at least a couple of others.

Amit, There isn't necessarily a right or wrong answer as to how many trades is optimal other than - * If you haven't already factored in transaction costs then you might give preference to fewer trades * If you are not fully automated then you may once again give preference to fewer trades *...

Amit - One thing that is very curious about the chart you have posted is that it once the price starts moving in a given direction (up or down) it tends to continue to move that way. To me that is an even more favourable strategy for that chart than trying to pick double tops etc.

I find that for shorter wavelenths (up to say 10 bars) this works well. However for longer cycles (10 to 40 bars) it's sweeter to get in a bit late and exit early. That is, only trade when the cycle is at its steepest. I tend to also use the conditional null * to selectively train nets...

Hi Mr Future, A really nice looking trading system. What appears particularly impressive about it is that you have trained it on a rising market and it still performs on a falling market. This is unusual but commendable. Just a few quick questions - 1) What do you mean 'two dimensional...

Hey Dommo, Yes, NST is Neuroshell Trader. It is a in the most part a very well designed and laid out program. There are also a variety of good add-ons available. Anyone can create a neural net in NST however to create robust ones does take a lot more perseverance. It really took me about...

It's tempting to think trading is a zero sum game but it's not. A private game of poker between a few friends is a good example of a zero sum game, ie the sum of the total wealth at the beginning is the same as the sum of the total wealth at the end. However the equities market is not like...

I don't trade with IG anymore. In the past two years I have had no issues with my current spread betting provider. They have been great and usually tighter spreads and rolls than IG. IG can take feeds from as many liquidity providers as they like - it's still hard to answer why you get...