intraday price action, reversion to mean considering also volume
Always with regard to doing some backtesting evaluations aimed to spotting reversal, let’s consider only intraday, and only slow Guppy Multiple Moving Average - GMMA.
reversals seem to be often signaled by the concurrence of:
1)high distance (say 1 or 2 STD, it should be checked out) from vwap, while
2) price vs all slow GMMA, or at least most tehm, are on opposite sides compared to vwap.
Then the problem is the determine when a backtesting exercise could take the signals of starting reversals.
Two possible evaluations can be done: (1+2)_a) or (1+2)_b)
(1+2)_a) In extreme movements one quick and dirty blind criteria could be that of taking the signal when the distance of price from vwap is greater than the distance of the slowest mov.avg on the other side of vwap; at this point enter and manage the trade in backtesting and wait its reversion to the mean.
(1+2)_b) in other conditions, not so extreme, the signal could be based (I haven’t yet completely figured out how ) both by looking to crossovers between price and fast GMMA or also, alternatively, at crossover between prices and a short length (say 3 or 4) McGinley Dynamic indicator.
Too bad I can’t program!! It would be great fun to carry out these kind of testing.
reversals seem to be often signaled by the concurrence of:
1)high distance (say 1 or 2 STD, it should be checked out) from vwap, while
2) price vs all slow GMMA, or at least most tehm, are on opposite sides compared to vwap.
Then the problem is the determine when a backtesting exercise could take the signals of starting reversals.
Two possible evaluations can be done: (1+2)_a) or (1+2)_b)
(1+2)_a) In extreme movements one quick and dirty blind criteria could be that of taking the signal when the distance of price from vwap is greater than the distance of the slowest mov.avg on the other side of vwap; at this point enter and manage the trade in backtesting and wait its reversion to the mean.
(1+2)_b) in other conditions, not so extreme, the signal could be based (I haven’t yet completely figured out how ) both by looking to crossovers between price and fast GMMA or also, alternatively, at crossover between prices and a short length (say 3 or 4) McGinley Dynamic indicator.
Too bad I can’t program!! It would be great fun to carry out these kind of testing.