Forex Automaton

During the month of June, the sixth month of live performance, the system went through a major upgrade. The upgrade is expected to further improve prediction quality for daily high and low (as measured by Pearson correlation coefficients between predicted and actual logarithmic returns) while its effect on the prediction quality for daily close is uncertain. This document consists of a summary section followed by 14 subsections, dedicated to the individual exchange rates tracked by the system. Those contain color-coded charts of the performance and details pertinent to the specific currency pairs. For comparison with the previous month, you may want to take a look at the May review.

Changes in the algorithm

Refer to the original announcement.

Figures of merit

The figure of merit used, Pearson correlation coefficient between the forecast and real logarithmic returns on a day scale, is a measure of forecasting quality.

By construction, the Pearson correlation coefficient is a quantity bounded between -1 (the forecast move and reality are total opposites) and 1 (the forecast is perfect). A success or lack thereof on every trading day makes a contribution to this quantity.

In order to make a large positive contribution, one needs a coincidence of a large move in a currency pair with a large forecast move in the same direction. Since a hypothetic rational operator of the system will not pursue small forecast moves, understanding this to be a noisy system, a forecast with large magnitude is more likely to result in a successful trade.

Why use Pearson correlation coefficient instead of running a model portfolio? This independent figure of merit, characterizing the forecasting component regardless of the money management strategy used, allows one to split the complex problem of trading system development into independent tractable pieces, and solve each pieces of the problem separately, eventually combining a winning forecasting system with a winning money management one, having an independent quality assurance process for each.

Pearson correlation coefficient for close could be interpreted as an average daily return (expressed as a fraction of price rather than fraction of capital) of a hypothetic portfolio where every trade is weighted according to the magnitude of the predicted return, if the returns had not been logarithmic. 

Performance tables

In the tables, data for the "month" column are taken from the "month" column of the last day's forecast, which reports 24-trading-day running average of the Pearson correlation figure of merit. End of June 2010 data are taken from the June 30 update. End of May 2010 data are taken from the May 31 update.

This month's effect on the performance since inception was, as Table 0.1 indicates, positive for EUR/USD, USD/CAD, USD/CHF, USD/JPY, EUR/AUD, and EUR/CHF (6 pairs in total) -- their performance since inception improved compared to end of May. It must be noted that the "month" column is for rough orientation only as it gives the average over 24 most recent trading days which may not necessarily be the month. In contrast, the conclusions on the basis of the "since inception" columns are strict. The mean over markets for the Pearson correlation coefficient for the month was negative.

In all three tables, the precisions of the mean "since inception" improved (uncertainties went down) compared to May. It is remarkable that at the same time, the mean values for low, high and close went up. For day low and high, we again attribute the improvement to the v1.0 upgrade performed this month and the v0.5 upgrade performed in January. Negative performance for close continues; since the live launch in late December 2009, February 2010 remains the only month of solid positive performance for close.

The figures of merit for the low and high play the role of a canary in a coal mine, ruling out the possibility of some of the less subtle problems. The continued (indeed, improving) high quality of daily low and high forecasts indicates that the canary is alive.

Usage strategies

Continued lackluster performance of forecasting for daily close makes us switch research attention to strategies relying more heavily on the stably high forecasting quality for daily high and low, and not so heavily on daily close. The break-through of the month in this direction was the realization that in order to use such a strategy (Day Range Strategy for short) effectively, one will need to use a forecasting regime (defined by the value of an adjustable parameter nicknamed Fred in our reports) which is different from the maximum of the Pearson correlation coefficients. A decision has been made to let Danica be the system optimized for the high predictability of daily extremes in terms of correlation coefficients, leaving the question of an optimal trading strategy for this forecasting regime open. Another system with a definite usage strategy (Day Range Strategy referred to above) using the daily high and low, in combination with their predicted directions for the day, as stop loss, profit target and direction for the day, is about to be launched in parallel to Danica.

System performance in the individual markets

The following 14 subsections are dedicated to the specific currency pairs, useing a particular method of charting to illustrade performance. Two charts will be presented for each. The first chart will show all trade ideas. This corresponds to trading in every pair every day.

The second chart will highlight the days when forecasts for all three components of a day's candlestick were pointing in the same direction. I call this level zero (L0) requirement or L0 trigger.

Table 0.4 shows little or no effect on the quality of trades from the L0 trigger this month, as judged by the simple tally of wins and losses. Note that such a simple counting approach ignores the problem of the relative impact of these wins and losses. Therefore it can only serve as an illustration to supplement more quantitative studies.