
	News
	Programmer wanted We're looking for a skilled C++ prog- rammer [2011-08-20] New service We now offer neural network modeling [2011-07-15]

Backtesting

Are you trying out your own ideas or developing your own trading systems, and are curious of the performance of the system? We can help you to backtest your strategy and create performance reports, to make sure that your system is robust enough or to give your customers statistical materials if your system is for commercial purpose.

We have access to historical data and can backtest your strategy on:

The majority of stock exchanges since 1980
The majority of stock indices
16 currency pairs (1 minute data or tick data)

When the backtesting is finished, we can either provide you with a thoroughly performance report or just a simple summary. This includes all important ratios of the strategy's performance, both as a summary and for each individual market, which gives you a good foundation for further analysis.

Furthermore you have the possibility to optimize some parameters. We can help you with the analysis of the strategy's performance, such as MAE-analysis, drawdown and probability models for different scenarios to occur.

Tick data

For the major forex pairs, we offer testing on tick data. Why would anyone want to use actual tick data for their backtests? When running backtests in Metatrader the platform use 1m data to create interpolated ticks. This means that the ticks used is not the actual ticks traded, but instead made up ticks based on Open, High, Low and Close of that bar. Even if it seems like 1m data can be enough, especially when you trade on higher timeframes, it is always recommended to verify the results on actual tick data.

The following tests are from a scalping technique. The first graph shows an exceptional result, based on 1m interpolated data. The equity curve is smooth with a steady incline.

The second graph however is the same strategy with the same settings and the same time period, but tested on tick data rather than interpolated 1m data. This result reflects what actually happened during the bar, and how that affected the results. This shows that we can’t have the same expectancy of the strategy as we first thought, and by verifying the previous test results on tick data can save you a lot of money by avoiding bad strategies.

The tick data we offer range back to 2002-2003, depending on which pair. Currently 11 pairs are available (AUDJPY, AUDUSD, EURCHF, EURGBP, EURJPY, EURUSD, GBPJPY, GBPUSD, USDCAD, USDCHF, USDJPY).

If you need to test your strategy on tick data, please do not hesitate to contact us for a free quote. We use 3 multiprocessor backtester machines that will run the tests quickly and efficient. The cost for tick data backtesting is 5 EUR + VAT per hour, with a setup fee for each symbol. When you request a quote, please attach the strategy you want to backtest so we can evaluate the time the tests will take.


	Testimonials
	I just received the final EA from System Investors. To my great satisfaction they delivered ahead of schedule. During the time of making the EA they has always been very punctual, kept thier promises and fast to respond. I my self does not speak the language of coding at all, so we have had some misunderstandings during the process. But we found a solution that worked for both of us. I have no worries sending more projects to System Investors, so all in all very pleased. Kevin Rytter, Denmark I been working on my own strategies for some years, with various results. I took the chance to let System Investors try to improve one of my strategies, and I've now been profitable 9 out of 10 months since then. I can definitely recommend System Investors! David Ljungquist, Sweden Read more testimonials