Dynamic Stop-Loss

In order to limit the losses many traders employ a stop loss strategy. The basic idea is that when a stock or an index is falling, it might continue to fall. In order to be protected against unlimited losses, a trader may choose to close the position after having incured a certain loss and enter again at a later time.

A dynamic stop-loss strategy increases the stop-loss trigger over time when the stock price increases. In our case the stop-loss trigger is always relative to the peak of the index since we bought it.

I have investigated a simple dynamic stop-loss strategy in selling the Dow Jones Industrial (January 1930 until October 2008), whenever it has dropped a certain percentage x% since the peak value since we've bought it. We then enter again y days later.

The best results of more than 630 different combinations I tested was achieved when we sell the index whenever it has dropped by 30% from where we bought it and buy again 1 day later. Compared to a simple buy-and-hold strategy this leads to very slight an underperformance even when we ignore transaction costs.

The ceiling depects a buy-and-hold strategy and the lower mesh shows the stop-loss-strategy.

As visible in the mesh diagram above, the ceiling depicting a simple buy-and-hold strategy where we achieve a bit more than 6.3% p.a. For the stop-loss strategy it doesn't seem to make a big difference what kind of stop-loss we use. As long as we buy again as soon as possible we only underperform the buy-and-hold strategy very slightly.

The mesh diagram below shows what % of the trading days we end up being long depending on the strategy we apply.

If we perform the same analysis with a different index, such as the SMI, the result is completely different. With what we identified as the maximum for the Dow Jones, would lead to significant underperformance in the SMI.

We can conclude that a dynamic stop-loss does not contribute any value when applied to stock-index trading.