[…]Quality Before we get to the tests, I need to explain a new metric I will be using. At Connors Research we use Individual Trade Quality, ITQ, when we were comparing results of non-portfolio tests, such as these tests. The simple way to understand ITQ is it analogous to Sharpe Ratio in a portfolio test. To get more details on ITQ see How to Measure the Individual Trade Quality of Your Strategy. Connors Research has adopted a measure of risk-adjusted return per trade which uses concepts similar to those incorporated in the Sharpe Ratio, but translated to individual trades. We […]
[…]No commissions. Buy and sell on the last trading day of the month at the closing price. Rebalance all positions on this day too. All returns include dividends. Baseline tests Here we have the returns for the SPY. The Guggenheim S&P 500 Equal Weight ETF, RSP, all stocks are approximately .2% in size. Then equal weight stocks using data from Norgate Data which has historical index constituent information. All the S&P500 stocks as equal weight with a rebalance each month. The results are close to RSP and with the yearly numbers, which you can see in the spreadsheet. The weightings […]
[…]continue to paper trade the strategy until it gets above the moving average and then trade it live again. The general idea being that you get out when the strategy is doing poorly and get back in when it is doing well. Also once a strategy breaks, this gives you a simple way of getting out of it. In this example we would stop trading during the blue oval because the equity curve is below the moving average. Often people will pick an indicator to use and then trade the equity curve live without seeing how the backtested results may […]
[…]market conditions (e.g. five up days in the SPX)? If so, maybe this is the start of a viable, standalone trading strategy. As indicated, this post builds on the previous post, so please refer as needed. Baseline Our research is on S&P 500 stocks from Jan 2001 to Aug 2013. We will focus on the five day return, because that time frame is what looked most promising in the initial research. A look at every S&P 500 stock, every day, and calculating the five day return (close to close), results in the following baseline stats. Exactly N Up Days Using […]
[…]for my tests I primarily use AmiBroker. I sometimes use Excel and MySQL. Where do you get your data? Currently, I subscribe and recommend Norgate Data. In particular the Platinum level. Is the data survivorship bias free? Yes. Are dividends and capital gains included? The data has been adjusted for dividends and capital gains.. Can you share your stock data? I cannot per my license agreement with Norgate Data. Is your S&P500 stock data as it existed or only a snapshot of today? The data is it existed in the past. No pre-inclusion bias. Do you include commissions in your results? […]
[…]a trade like that is psychologically hard to do. The third worst trade was down 53% over 21 days. Again, these are hard to handle and can cause one to stop trading. This caused me to abandon the strategy several years ago. Max Loss Stop I was discussing this strategy recently at a NWTTA meetup and of course the discussion came to stops. What if you simply exit when your end day loss is more than X%? I just love this test. As you can see no matter what value you pick for your max loss stop your CAR and/or […]
[…]ideas we will investigate are: Monthly rotation (instead of quarterly) Using an additional filter to make sure the stock is healthy. These include Close above 200 day moving average Close above 50 day moving average 50 day moving average above 200 day moving average Stock return over look back period is positive In the last 10 days the stock has made a 1 month high In the last 10 days the stock has made a 3 month high Maximum loss stops The Base Case The test range will be from 1/1/2004 to 12/312013, giving us 10 years of data. The […]
In Tactical Asset Allocation (TAA) or Dual Momentum (DM) strategies, they often will use trend-following or momentum to decide whether to invest in asset or not. I have two questions. One, how often does either trend-following or momentum out-perform buy and hold? Two, of the two which one out-performs the other more often? Trend-Following Rules Buy Last day of month Close is greater than the 10-month moving average Buy on open of next day Sell Last day of month Close is less than the 10-month moving average Sell on open of next day Momentum Rules Buy Last day of month […]
I will be speaking at the 2015 Australian Technical Analysts Association on May 15 to 17, 2015. My topics are “The development of an S&P500 stock weekly rotation strategy” and “From Internet Article to Trading Strategy: An ETF Monthly Rotation Strategy.” For more information about the conference go here. I am excited to meet some of my readers at the conference. I will have two free days to explore Sydney before the conference. If you have any suggestions on things to do and see (I will not have a car) or places to eat (from hole-in-the-wall to fancy) or hotel […]
[…]trading strategy? Improve a trading strategy? Validate a trading strategy? My 18 years of professional experience in designing, coding, testing, and validating trading systems is at your disposal. I specialize in quantitative research. If you can describe rules to your trading strategy, I can code it. I can assist you in the development of your strategy along with evaluating your strategy using common metrics or custom metrics that you have. I can help validate your existing strategy. This can give you peace of mind that no coding errors exist in your strategy. Strategy validation is a very important, but neglected, […]
