Quote from Kohanz:
Rol,
Good to see things are still going well! You mentioned earlier in this thread your general method of coming up with this strat was to look through the historical data for when a stock had a large daily uptick, then look at the daily bars prior to that and see if you can observe a quantifiable pattern and then to automate that.
I'm curious if you took news into account in your strat development to exclude bumps that occurred due to news. I'd love to hear why or why not you included or excluded that data.
Thanks!
Kohanz,
That is interesting that you mention factoring news into the equation, because I have considered that. My back testing does not regard news, and I imagine it would be very difficult if even possible to address news historically to back test. However, I have made some observations about how news relates to stock price. Generally upside earnings surprises are made public after markets are closed, so they generally gap the next day and you miss a good part of the price move. While it appears to me that downgrades are made public around the market open, often leading to steep sell offs right after the news is made public. It would be nice to have a way for the strat to avoid going long stocks that had the word âdowngradeâ, in their recent news. TradeStation does have a way to incorporate news into a strat, but I havenât looked into this yet. I believe many professional traders actually have trading algos that use news like this, which would help explain sudden rises in volume within seconds of a news release. A nice counter strat to what I currently am doing would be to short stocks that have, for example, âdowngradeâ or âearnings missâ in their news, accompanied with volume spikes. I have decided that when news is really bad for a stock, whether it has merit or not, to fight the trend is generally not profitable.
