Our testing has confirmed a tiny edge in some price action approaches. Here is one we published back in 2008: Where are the studies and research from the jokers who make claims like 90% accuracy and such? How about the evidence-- I present ours--- Once in motion, an object will remain in motion unless acted upon by an outside force. This basic principle of physics can be applied to the stock market. Price will fall until it reaches the point where investors consider it a bargain. At this time, the force of the bargain buyers will reverse the fall by applying the force of buying which in turn causes shares to climb. This is the one constant of the stock market. Weakness is bought and strength is sold. The question has always been, just how weak is weak enough to trigger the buying? In addition, what is the fine line that could indicate the stock will not attract buyers regardless? We conducted extensive research to quantify this buying the dips observation so that it could be used to provide a trading edge. What we found was quite astonishing. Here is the results of the study, remember, all the names were trading above their 200 day Simple Moving Average. This moving average is the fine line that differentiates stocks that will likely bounce to those that potentially will continue the down trend. Stocks that closed down exactly three consecutive days, on average, outperformed the benchmark 1-week later (+0.36%). Stocks that closed down exactly four consecutive days, on average, outperformed the benchmark 1-week later (+0.54%). Stocks that closed down exactly five consecutive days, on average, outperformed the benchmark 1-week later (+0.63%). Stocks that closed down exactly six consecutive days, on average, outperformed the benchmark 1-week later (+0.82%). Stocks that closed down exactly seven consecutive days, on average, outperformed the benchmark 1-week later (+1.06%). As you can see, there is an edge in buying stocks that have fallen 5 or more days in a row.
Ok. This research was completed on one of the world's largest data base of trades-- where do you suggest additional data for testing be obtained to get deeper under the surface? thanks. surf
Plenty of intraday SPY and ES data available....lots of repetition in these mkts with regards to them taking out new highs/lows intraday, the speed at which these levels are taken out (or lack of speed), the distance price moves off a low or high and the PA that follows these moves (sniffing out whether dip buyers (or rally sellers) have the momentum to continue a trend or relent it altogether. Get creative
The conclusion, may I say, is incorrect. =) I think the correct version is, There is an edge in buying every single stocks that have fallen 5 or more days in a row with equal bet size. Even failing to buy one single stock from the set, at the exact closing price (which is not quite possible anyway) and also exiting the position at exactly n-day after at another exact closing price (again, a task not quite possible), will not give you the edge you are talking about. That does not take into account of your dataset having survivor bias issues. Can't help putting on my statistician hat.
The studies were strictly with the stock market, not index futures. In fact, the extensive and exhaustive academic and practical research I have been privy to regarding the emini markets indicate no such correlations particullarly on short term basis. It's mostly noise, it's best to deal with it other than seeing pretend patterns etc. Am I correct in surmising that your method is highly subjective and continually adapts to the current market? Reading between the lines this appears accurate.
This "edge" looks exactly like the edge found by an statistician who drowned in a lake of average depth 6 inches.
"pretend patterns" LOL!!! You are not correct in your summation, I am just giving you SOMETHING to start from. It would be up to YOU to come up with something objective. Can't spoon feed the masses here