How It Works

The main goal of Forex-Forecasting.com is to provide daily and intra-day high-quality predictions of FOREX market prices. You will receive currency predictions in a table/graphics format, based on five different time frames period (5 and 15 minutes, 1 hour, 1 day, and 1 month) along with the appropriate buy/sell signals.

Forex-Forecasting.com uses prediction algorithms initially developed for www.stock-forecasting.com and adapted for the Forex market.

Services:

You can access our service in two ways:

• Online through our website.
  Once you have created an account, you will receive access to the control panel and the tools we have to offer.
• Using your favorite software, such as Metastock, Metatrader, and others.
  You'll need to download and install the Forex-Forecasting plugin. Then, you can use our predictions in conjunction with the formulas and trading algorithms you already use.

The Concept:

Many samples of currency trends have a wave (or a non-periodic, oscillating) structure. This can be mathematically represented as a combination of the number of harmonics with unknown, changing frequencies and the trends’ amplitudes. Thus, information about these harmonics is very useful for both time-series forecasts (market price predictions) and decision support (buy/sell advice). However, ordinary analytical methods cannot be used to separate the variable parameter harmonics.

We have developed a special forecasting method for economic time-series based on our innovative and unique technology. Central to our method is the decomposition of both the trend and oscillatory components of the time-series with the help of digital filters. This special adaptive technique, based on neural networking, is used to update our models and detect days when the price time-series changes its properties (our "know-how").

In contrast to other methods, our technique can identify long-term trends and oscillations with changing frequencies while delivering much more convenient results than, for example, a Fourier analysis.

You can read more about our method here:
Adaptive Harmonic Component Detection
Blind Deconvolution and Separation Signal Processing Adaptive Forecasting of Wave Non-Periodic Time Series