The Risk of Trading: Mastering the Most Important Element in Financial Speculation

CRITERIA FOR DETERMINING EDGE

The greater the quantity of our sample size, the better chance we have of accurately determining our edge. Sample sizing is a powerful component to determining a probable outcome and is an essential to any valid back-testing plan. The law of large numbers goal is not intended to precisely predict a future outcome but merely to reduce the differential of the expected outcome vs. the actual one. Just because the ever-changing variables of the market are at work, providing a variance that even the best traders cannot detect, we can use a valid sample size to aid in detecting edge. In order for a historical sample to be considered eligible for possible detection of edge, we must be sure our historical tests meet the following criteria:

1. The historical trades used in the sample have occurred under nearly identical market conditions. Since market dynamics are ever-changing, this places strict limits on how accurate historical trade data can truly be. As risk managers, we can take trade activity during “normal” markets and optimize the ability to help predict future outcomes. We can eliminate trades that occurred during announced news events or other times of extreme market volatility unless you are looking to establish a historical outcome using such parameters. In this example, you can create a sample from trades that only occur during news events. In this case, the news event is the normal market condition that we are seeking. On the flip side, extracting ...