Bayesian Statistics in Trading
In our class, we have seen Bayes Theorem in calculating conditional probability P(A | B), which is the probability of event A happening given that event B has already occurred. This means that Bayes Theorem gives you the probability of a future event occurring based on prior events, which is why it’s so popular and has its own field within statistics. The article attached below looks into how this Bayesian Statistics is used to make a profit in trading, which makes sense given that the whole point of trading is to earn money, and knowing the probabilities of future events would definitely come in handy.
The crucial assumption here is that the probabilities are constantly changing since trades are occurring very quickly when the markets are open. So, the probabilities must constantly get updated for the forecast to be accurate. There are complex trading strategies that are based on Bayesian Statistics, but we will not be discussing those in this article. However, we will look over the basic principles of how to trade given the probabilities can be computed in real-time to keep up with the current state. In the beginning, an educated guess is made to kick off the trades, and this shouldn’t be just a random guess, it should be based on the current trends. After that, we follow the following four trade update rules
- If your stop loss is triggered you exit with a small loss
- If the trade moves in your favor you let it go on
- For a big winner, you move your stop loss to a trailing stop to maximize the win
- If your profit target is hit you lock in profits as the risk/reward may become skewed against you
All these update rules follow one principle: “If you’re earning money, maximize profit, if you’re losing money, minimize loss”, which will all go towards helping the trader make money.
Original article