Forex Automaton

Our primary goal is to create a public information service providing financial markets forecasts, based on our proprietary forecasting tools: an automated trading system -- a Forex Automaton™. A live forecasting system operating on daily time scale, Danica, is online since late December 2009. An hour-scale forecasting system, Heidi, and a daily signal EUR/USD system running a model portfolio, Demi, were launched in August 2010.

Our secondary goal is to discover, quantify and monitor the very existence of sustainable opportunities for profit-making via systematic trading. Or simply put, to monitor the degree to which these markets are more predictable than a "fair game" -- to a trader without access to insider information.

Four years of ForexAutomaton's life are over. This year I have been busy starting a currency risk consulting company, along with two partners, therefore less than the usual amount of research content has been published here. Nonetheless, the automated systems -- Danica, Demi, and Heidi -- have been living their silicon-based lives as before, with little to no interference from their creator. This fourth annual report is dedicated entirely to news from their side.

April 2012 was the first month of the upgraded Danica 1.3. While the forecasts for daily low and high were strongly positively correlated with reality, the predicted changes in the daily close were on average anti-correlated with the actual ones.

This system performance review consists of a summary section reporting the figures of merit for the forecasting quality, followed by 14 subsections, dedicated to the individual exchange rates tracked by the system. Those contain the usual green-yellow-blue-red color-coded charts of the performance, such as shown below, for each currency pair.

For comparison with the previous month, you may want to take a look at the March 2012 performance review.

This monthly forex correlation analysis report is computer-generated. The report follows structure and definitions explained in a separate document, which can be used as a user manual. The report deals with the following exchange rate time series: AUDJPY, AUDUSD, CHFJPY, EURAUD, EURCHF, EURGBP, EURJPY, EURUSD, GBPCHF, GBPJPY, GBPUSD, USDCAD, USDCHF, USDJPY.

During March 2012, the third month of the third year of Danica's live operation, the system continued on auto-pilot without parameter changes. While the forecasts for daily low and high were strongly positively correlated with reality, the predicted changes in the daily close were on average anti-correlated with the actual ones.

This system performance review consists of a summary section reporting the figures of merit for the forecasting quality, followed by 14 subsections, dedicated to the individual exchange rates tracked by the system. Those contain the usual green-yellow-blue-red color-coded charts of the performance, such as shown below, for each currency pair.

For comparison with the previous month, you may want to take a look at the February 2012 performance review.

This monthly forex correlation analysis report is computer-generated. The report follows structure and definitions explained in a separate document, which can be used as a user manual. The report deals with the following exchange rate time series: AUDJPY, AUDUSD, CHFJPY, EURAUD, EURCHF, EURGBP, EURJPY, EURUSD, GBPCHF, GBPJPY, GBPUSD, USDCAD, USDCHF, USDJPY.

This upgrade is affecting the way forecasts are generated, in certain situations. In the previous versions, when the predicted close comes out to be above the predicted high, the system would not touch the high, but instead would adjust the close so that the adjusted close is half way between the original one and the high. Similar algorithm would be applied when the predicted close comes out to be below the predicted low.  

In version 1.3, in this situation, we pull down the high so that it equals the close, or we pull up the low so that it equals the close.

The new way is more consistent with my understanding of relatively high prediction quality of high and low as a mundane phenomenon related to continuity of price in the diffusion process, as explained. There is no reason to treat information in the predicted close as secondary priority to that in the extremes, yet that is what took place before this upgrade.

During the past 10 years, liquidity in the FX markets is known to have been growing. One might expect that various non-commercial participants, including those with capabilities to research, fund and execute systematic and algorithmic trading strategies, exhaust the alpha-generating potential of this market and drive it towards efficiency. Do spot foreign exchange markets really evolve towards efficiency? I use the ForexAutomaton CERPI 1H.1M inefficiency index (hourly scale, monthly data accumulation period) to look at the trend of the past 10 years (2003-2012), and compare the picture with that of the evolution in the instantaneous ("non-predictive") correlation strength (CERCSI, pronounced "Sir-see") and volatility.

  Currency Exchange Rate Correlation Strength Index (CERCSI) is a measure of zero-lag correlation strength with a specified timeframe (time scale) and measurement interval. For quantities we report regularly, measurement interval is the same as period of measurement. The quantities reported in the Forex Correlation Analysis Reports 1H.1M section of the site are measured monthly at the end of a month and include the whole month of data. This article defines the way CERCSI is constructed.

Currency Exchange Rate Predictability Index (CERPI) is a measure of market inefficiency with a specified time frame (time scale) and measurement interval. For quantities we report regularly, measurement interval is the same as period of measurement. The quantities reported in the Forex Correlation Analysis Reports 1H.1M section of the site are measured monthly at the end of a month and include the whole month of data. This article defines the way CERPI is constructed.

Forex Automaton is launching a new type of information product: a monthly correlation analysis report. The reports are computer-generated and follow a pre-defined format. This document explains the purpose and structure of the reports and is intended to serve as a brief user manual to the new information product. It also collects in one place any other information necessary to understand the reports, which otherwise would have to be repeated in every report. The document explains the quantities shown in the plots and relationships between the plots.

Data accumulation for the present incarnation of Heidi began on December 1, 2010 and the first predictions (in back-testing mode) were generated for March 16, 2011. The present system was announced and began live operation on May 23. Now, eight months later, I take a look at the first statistically significant performance figures of merit and make an adjustment to one of the system parameters.

This post begins monthly peformance reviews for the Demi line of EUR/USD daily time frame trading systems. There are four systems in total, differing only according to the choice of the closing hour of the "day": 3, 5, 9 and 11 am Eastern time. The systems went live on August 25, 2010, after extensive back-testing. Since then, the parameters found in the course of the back-testing have been frozen. Even though the system output (see Demi section of the site) is available only to the registered members, the performance reviews will be available publicly.

For clues as to what effects to expect in forex rates from the consequences of the March 11th earthquake in Japan, I look at historical data from 1995, the year of the Great Hanshin earthquake. Great Hanshin earthquake occurred on Tuesday, January 17, 1995, at 05:46 JST (16 January at 20:46 UTC). There are at least two topics to explore. We look for quantitative effects of repatriation of funds into Japan to fund the recovery efforts on the exchange rate of the Japanese Yen to the US Dollar. Another possible concern is the carry trade implications.

Hourly time scale in forex provides roughly 24 times more statistics, than the daily scale. Therefore, if a single adjustable parameter for all currency pairs is justifiable in the daily system optimization (Danica), 24 parameters can be adjusted for Heidi without jeopardizing significance of the results. The natural way of increasing the number of parameters is to split the 24-hour day into time window "seasons" and treat the seasons as independent pattern recognition problems in the code and as independent optimization problems on the stage of optimization.

Intra-day seasonality in JPY is researched by averaging hourly logarithmic returns, grouped systematically into temporal "bins" according to the time of the day. Two types of average, the average across years of observation for each instrument, and an average across instruments for each year, are presented. The instruments are USD/JPY and four popular crosses involving AUD, EUR, GBP and CHF. The effects look practically insignificant given the typical levels of trading costs available to a retail trader, and given the lack of time stability.