
Euro FX 5 minutes bar, today and yesterday
In Blue the bollinger band calculated with the average price (high + low / 2) at 89 periods.
The red bands represent my method to filter the data with a centered T3 Filter with 89 periods (it uses future data to erase the lag, so the last n-bar in blue are forecasted with autoregression algorithm and might change)
More Details about the filter i use:
T3 is a high order low pass denoiser made of six-pole nonlinear Kalman filter . The Kalman filter is an efficient recursive filter that estimates the state of a linear dynamic system from a series of noisy measurements (from wiki)
Inputs of the Filter:
Periods T3 = Periods for Tillson's T3 filter
Slope T3 = Slope for Tillson's T3 filter (0.7 to 0.83 for usual value) (Higher value mean more dynamic filter) i use 0.7 that is perfect and fit better the data.
Slope T3 = Slope for Tillson's T3 filter (0.7 to 0.83 for usual value) (Higher value mean more dynamic filter) i use 0.7 that is perfect and fit better the data.