Find out how our AI Trading works in Forex trading.
Economics + Mathematics + Artificial Intelligence.
What makes our AI trading solution unique is the experience of our team of experts in economics and mathematics, for developing their own trading strategies and application into AI forex trading systems.
Our team consists of professionals with over 19 years of experience in the Financial Market, Doctors in Advanced Mathematics and Artificial Intelligence.
Technical + Fundamental + Sentiment Analysis
Our Forex trades are emitted by a combination of different AI systems, with more than 100 features, to define 40 different strategies which simultaneously consider:
Our AI analyzes the fundamental, technical, and sentiment factors that affect forex trading, and all criteria are evaluated in real-time using different weights. After completing the analysis, the choice of the signal emission strategy is made considering the increased likelihood of short-term profit and the amount available for investment.
Using technical analysis, our algorithm considers not just the patterns of various trends, including support and resistance levels and cross-indicators, but WiseSignals is also able to create its own real-time index for each currency pair, that is used to identify what is the best direction and target.
In relation to fundamental analysis, the AI makes a correlation within the economic calendar to find news data that can affect specific currency pairs. In doing so, the algorithm is able to identify the following scenarios:
Market inactivity to not send signals.
Avoidance of certain risky entries.
Measure favorable patterns after-events.
AI trading performance (Win Rate)
The Win Rate of our trades varies and can reach up to 95%. The performance of our AI trades are based on efficient frontier, which means that our model is profitable whenever the Win Rate is greater than 60%. Our strategies seek stable returns and sufficient results to achieve an average return of 40% per year (APY), without the application of high leverage.
We know that what matters is the final result. Therefore, the investor needs to note that Win Rate is not always the best metric, as other signal providers disclose. For example, out of 100 trades, it is possible that only 1 trade is enough loss to make the 99 trades not profitable. In this case, the Win Rate would be 99%, but the final result would be unfeasible (negative). Therefore, we focus on the return of 40% of APY and not just on Win Rate.
Our strategy also considers that the performance must be composed of a final result in a monetary unit, such as Dollars or Euro. For this reason, although we make results disclosures in Pips, we always highlight them in Dollar. Furthermore, results in Pips do not always mean viable returns.
All our trades are placed in live accounts with our own capital and have been tested on more than 15,000 operations. We also calculate the annualized return (APY) and consider the benchmarking of the main exchanges for calculating the Sharpe Ratio. Our performance can be checked in real-time in our track-record clicking here.
We are the only platform that provides AI Forex trading solutions through AI with risk management for 28 currency pairs in real-time. With us, you can trade automated with AI-Trading in partner brokers or manually with AI-Signals in the broker of your choice.
Each trade of our AI has a target of profit between 0.10% and 0.65% and an average risk exposure between 0.15% and 0.65%. For example, if the capital available for investment is $ 1,000, each trade will allocate around $ 6 of your capital. Thus, in the case of closing with a profit, the total capital will be $ 1,006. However, if closed in a loss, the capital will be $ 994.
Our risk management also applies an exposure mitigation strategy and uses Risk-Adjusted Return Ratios such as the Calmar Ratio and its variants.
Live Track-Record and Backtesting
To define the set of 40 trading strategies, our AI ran
Our System is a true AI
More than automation, our algorithm is a multi-layer AI system, integrated with some of the largest forex brokers via API rest and FIX protocol. We use parameters that are constantly optimized to achieve the best returns, Sharpe Ratio, and Drawdown. We have a high precaution to avoid overfitting our models to any price data.
We use two main approaches to AI. The first one is a search of maximum return and minimum drawdown using a Multi-Objective Particle Swarm Optimization, which can validate our indicators by a walk-forward optimization method to simulate and move along the market as it was live trading in order to guarantee what are the best parameters for a certain encoding of indicators
In this way, we selected the set of indicators to use in our second AI, settling a time series classification via Convolutional Neural Network, which the architecture is similar to an AlexNet with data augmentation, based on our own math concepts. This second AI indicates if it is the buy, sell, or hold, for given security.
Our most effective indicators are made by ourselves, primarily developed in Python as a real-time algorithm and based on out-of-sample backtesting. To evaluate our model, we chose the MQL4 language to demonstrate our results on the MetaTrader 4 (MT4) platform. Our main libraries are Matplotlib, NumPy, Pandas, Parallelism with ray remote, and Talib.
Understanding our AITrading Methodology
Each of our signals consists of a Currency Pair (Symbol), Direction (Buy or Sell), Volume to Invest (Position Size), Take Profit and Stop Loss, for trading in forex live-trading, especially for retail investors. For live trading, it is necessary that the investor set its amount available and the type of contract used by the broker.
Decisions about the emission of signals are made exclusively by our AI and do not depend on human interaction. Whenever we identify a trading opportunity, our AI also sends real-time alerts by email and Telegram.
Due to our price range strategy, the signals can be used to invest whenever they are available on our website panel, regardless of the price. So, to avoid timing errors, the signals need to be copied directly from the dashboard.
When programming in non-application specific codes, such as Python, Wiseinvest does not depend on third-party applications to operate trading strategies and calculate the signals.