[…]his wishes and not share it. Given we could not remember the exit, I was forced to test several common exits I used back then. Testing dates are from 1/1/2007 to 12/31/2019. Why did I not include 2020? They skewed the results too much. As you will see in the next post. The Exits These are the exits I tested. 2-period RSI greater than 30 2-period RSI greater than 50 2-period RSI greater than 70 ConnorsRSI greater than 50 ConnorsRSI greater than 70 Close above the 50-day moving average The Results These results really surprised me. What I would call […]
[…]pair of fresh eyes looking for mistakes. This can sometimes take several iterations before I feel comfortable with the results. Closing Comments I fear the day when I make a post with an error in it. It cannot be avoided. It almost happened with the “DTAYS Post.” It took me several more iterations than normal to get it right. I was getting results that looked good but my gut said no. I came within a day of publishing the results before finding my error. Whether you do our own backtesting or have someone else do them for you or […]
[…]we have a set up, then enter a limit order for the next day at 5% below the close. Order good for one day only. Only place enough orders so if they are all filled you are not in over 10 positions If have multiple set ups, then rank from high to low by the 100-day historical volatility. Exit Rules Two period RSI is greater than 50 or after 10 trading days Exit on next open Simple mean reversion exit of waiting for the bounce. Base Results These are the results we will compare against. The Limit Exit The exit […]
[…]below the 200-day MA for 10 or more days. This new method has a lot less trades and whipsaws when compared to trading the 200-day MA cross. Bull Market Results Buy and Hold greatly outperformed everything as expected. If you trade these type of strategies, you must remember that during this time you will likely underperform. But comparing this sector strategy to my SPY strategy baseline it compares well. With CARs about the same but with a lower drawdown. Bull & Bear Market Results Now we can see why one would trade these types of strategies. Even though the […]
[…]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? Universe Size My first question was how many stocks are in the index? As you can see this is a small index with the number of stocks varying between 40 and 65. Few stocks have a rising dividend history. Misc. First, the first test is from 1/1/2007 to 8/31/2018 as in the original post. Buy Rules It is the last trading day of the month […]
[…]to February 19, 2020 Looking back from the market top, February 19, I calculated 33-day correlation for all stocks in the S&P100. There are 5050 pairwise correlations. Of those how many had a correlation above .70, which is typically considered strongly correlated? Only 100 or 2%. Lots of yellow and some reds in the chart below. S&P100: February 20, 2020 to March 24, 2020 On March 24, I calculated the 24-day correlation of all the stocks. Of those 5050 pairs, 75.5% had correlations over .70. WOW! Now that is taking a huge step towards a correlation of one. I did […]
[…]find most people use ‘gap’ to mean both a ‘lap’ or a ‘gap.’ I will use this definition for a gap. The Base Strategy These rules were chosen to have lots of trades and a large drawdown. I want to see how adding the gap rules helps these metrics. This is not a strategy I would trade given the large drawdown and the small average % profit/loss per trade. But maybe you like these numbers. Setup Rules Stock is a member or was a member of the Russell 3000 The Dollar Volume of stocks is greater than $500,000 Close […]
[…]platform used: AmiBroker. Data provider:Norgate Data (referral link) Good Quant Trading, Fill in for free […]
[…]platform used: AmiBroker. Data provider: Norgate Data (referral link) Good quant trading, Fill in for free […]
[…] My first random thought was what happens with different start dates. Meaning I had no good reason for picking that date. I could have just as easily picked some other date. Let us see what kind of results we get with two other dates. Correlation Here are the original results from 2000. I have also added the correlation of 2000-2012 vs 2013-2015. We can see the correlation of .74. So it appears that decile ranking in 2000-2012 is doing a good job for predicting the results in 2013-2015. What happens if we start in 2003? That would give us 10 […]
