Category Archives for "Stocks"
In my previous post, Adding candlesticks to mean reversion setup, we looked at how various candle patterns could help individual trades. Now we will see how those results translate to a portfolio. And why I usually only do portfolio level testing.
Should you avoid trades that have recently gapped? What if you are trading a mean reversion strategy and a stock has recently had a large gap? Is that a good trade to take? Avoid? Does it depend on the direction of the gap? I did research on this about 15 years ago. Let’s see what the current research says.
Recently, I have been working on a strategy that trades stocks with low dollar turnover. The initial performance was attractive and I was liking the strategy. But there were two issues that I needed to deal with in the backtesting. How much slippage to add to these stocks. The strategy enters and exits on the open and while looking over the trade list, I noticed some trades entered at the low of the day and exited at the high of the day. From my trading, I knew this would not be a realistic price. Should these cases get extra slippage? What follows is how I try to account for these issues.
I had a long-time reader, Cristian Franchi, send me a mean-reversion strategy that he wanted me to test and write about. What caught my attention was the rules differing from what I typically see and use. Different ways of measuring strength of a sell-off and volatility expansion. Along with a different type of exit being used on a mean reversion strategy. Not simply waiting for the bounce.
While doing research on a mean reversion strategy, I was really happy with the Compounded Annual Returns (CAR) of 51%. I was thinking, I may have a new strategy to add to my stable of trading of trading strategies. A big fact I liked was the strategy used no market regime filter.
Then I looked at the yearly returns. The 2020 return through July 31 as 444%! How much did the CAR depend on this year’s numbers?
Back in 2018, I wrote a post, Backtesting a Dividend Strategy, which was conceptually based on the S&P 500 Dividend Aristocrats. Just recently, Norgate Data started offering historical constituent data for the S&P 500 Dividend Aristocrats index. This would be a much ‘cleaner’ version compared to what I was trying to do in my original post. Would using this index produces better results?
There is a saying: “in bear markets correlations go to one.” I wanted to see how true that is for both stocks and a basket of ETFs. Now they don’t go to exactly one, not that I expected that, but they take some large steps towards one.
In my last post, Inverse Volatility Position Sizing, I tested inverse volatility sizing on a monthly rotation strategy. I saw very little difference in the rest results versus equal position sizing. I was talking to a trading friend about the research and how I was surprised at how there was not any difference in the results. He suggested creating an index using this method.
Now, this sounded like an idea with good potential. And even better it should be easy to test since I had the code written already.
Recently I’ve had several of my consulting clients come with a strategy that uses Inverse Volatility Position Sizing. The basic idea is that the more volatile positions have smaller size while the less volatile ones get a larger size. I have always been a fan of equal position sizing for several reasons. One, it is simple to do. Two, it is one less variable to optimize on and thus overfit on. Three, I rarely see much change in the metrics I care about when using more sophisticated algorithms.
Inverse Volatility Position Sizing is said to slightly reduce returns but has a big decrease in drawdowns and an increase in Sharpe Ratio. Time to test and see if that is true.
Most of us focus our research time looking to find better entries. We don’t spend enough time thinking about our exits. I am definitely guilty of this. A popular way to enter a mean reversion trade is by using a limit order. I use that on the strategy on RSI2 Strategy: Double returns with a simple rule change post.
The exit on that strategy is on the open. Many people don’t like exiting on the open because of the volatility and the belief that you will get a bad fill. What if we exit instead using limit orders? I tested this idea years ago. Time to revisit an old idea.