Page 76 - Ultimate Guide to Currency Trading

Finding Data for Your Program

The most basic of trading systems use past data to test the probability of future events. You can set up
a very simple system that can help you deter-mine the potential for a currency pair to move in an
either up-or-down direction using a combination of past closing prices of currency ETFs and economic
indicator ETFs. You can build a system such as this using information (called data points) off websites
such as Yahoo! Finance and Google Finance. Once you have the required information, you can arrange
the data points on MS Excel and perform simple regression analysis to build a model that can predict
with a high degree of accuracy the direction of your target currency pair, and to a lesser extent, the
percentage of that FX pair's movement.

The programming of MS Excel isn't hard, and the real value is the part where you find your
own data to fit into the model you are building. This means that although anyone can learn how to
build a system such as this, the .real challenge lies in finding combinations of data that have meaning
and influence in the direction and percentage movement in the target currency pair.

Regression analysis is usually the place where many hedge funds begin their numbers-
driven, quantitative-trading programs. These highly automated, highly tuned funds use
historical data, mathematics, and lightning-fast computers to create trading systems that
can be adjusted, updated, and refined continuously during the trading day.

To begin, you are trying to find which combinations of ETFs influence the target ETF. In
essence, you might ask your model how much the JPY ETF, SEK ETF, EUR ETF, S&P 500 ETF, VIX ETF,
Dollar Index ETF, TIP Index ETF, or any combination of other ETFs influences (and if so, by how much) a
target currency ETF. The first step in building any trading system using regression analysis is to find as
many factors that might possibly influence the target ETF or currency pair. In this case, the more the
merrier, as at your first run at spinning the wheel you will learn with precision which inputs have a say
in the direction and percentage movement of your trading target.

ETFs as Prepackaged Data Points

With this in mind, you should first determine your target currency pair, which will be represented by
proxy in the corresponding currency ETF. The second step is make a list of four to seven or even more
possible factors that influence the target currency pair. The key is to find proxy ETFs that track or
duplicate the direction and percentage movement of that potential influencing factor. You know that
fundamental, economic, and technical indicators influence the direction of a currency pair. Many of
these indicators are pre-packaged in ETF investment vehicles for the public to track and trade. With a
little research (by searching "ETF list" in Google), you will find listings of many ETFs that track
indicators that you can use as inputs in your model. There are ETFs for VIX, Inflation, T-bill prices, the