Dynamically Adjustable FilterIntroduction
Inspired from the Kalman filter this indicator aim to provide a good result in term of smoothness and reactivity while letting the user the option to increase/decrease smoothing.
Optimality And Dynamical Adjustment
This indicator is constructed in the same manner as many adaptive moving averages by using exponential averaging with a smoothing variable, this is described by :
x= x_1 + a(y - x_1)
where y is the input price (measurements) and a is the smoothing variable, with Kalman filters a is often replaced by K or Kalman Gain , this Gain is what adjust the estimate to the measurements. In the indicator K is calculated as follow :
K = Absolute Error of the estimate/(Absolute Error of the estimate + Measurements Dispersion * length)
The error of the estimate is just the absolute difference between the measurements and the estimate, the dispersion is the measurements standard deviation and length is a parameter controlling smoothness. K adjust to price volatility and try to provide a good estimate no matter the size of length . In order to increase reactivity the price input (measurements) has been summed with the estimate error.
Now this indicator use a fraction of what a Kalman filter use for its entire calculation, therefore the covariance update has been discarded as well as the extrapolation part.
About parameters length control the filter smoothness, the lag reduction option create more reactive results.
Conclusion
You can create smoothing variables for any adaptive indicator by using the : a/(a+b) form since this operation always return values between 0 and 1 as long as a and b are positive. Hope it help !
Thanks for reading !
Kalman
Well Rounded Moving AverageIntroduction
There are tons of filters, way to many, and some of them are redundant in the sense they produce the same results as others. The task to find an optimal filter is still a big challenge among technical analysis and engineering, a good filter is the Kalman filter who is one of the more precise filters out there. The optimal filter theorem state that : The optimal estimator has the form of a linear observer , this in short mean that an optimal filter must use measurements of the inputs and outputs, and this is what does the Kalman filter. I have tried myself to Kalman filters with more or less success as well as understanding optimality by studying Linear–quadratic–Gaussian control, i failed to get a complete understanding of those subjects but today i present a moving average filter (WRMA) constructed with all the knowledge i have in control theory and who aim to provide a very well response to market price, this mean low lag for fast decision timing and low overshoots for better precision.
Construction
An good filter must use information about its output, this is what exponential smoothing is about, simple exponential smoothing (EMA) is close to a simple moving average and can be defined as :
output = output(1) + α(input - output(1))
where α (alpha) is a smoothing constant, typically equal to 2/(Period+1) for the EMA.
This approach can be further developed by introducing more smoothing constants and output control (See double/triple exponential smoothing - alpha-beta filter) .
The moving average i propose will use only one smoothing constant, and is described as follow :
a = nz(a ) + alpha*nz(A )
b = nz(b ) + alpha*nz(B )
y = ema(a + b,p1)
A = src - y
B = src - ema(y,p2)
The filter is divided into two components a and b (more terms can add more control/effects if chosen well) , a adjust itself to the output error and is responsive while b is independent of the output and is mainly smoother, adding those components together create an output y , A is the output error and B is the error of an exponential moving average.
Comparison
There are a lot of low-lag filters out there, but the overshoots they induce in order to reduce lag is not a great effect. The first comparison is with a least square moving average, a moving average who fit a line in a price window of period length .
Lsma in blue and WRMA in red with both length = 100 . The lsma is a bit smoother but induce terrible overshoots
ZLMA in blue and WRMA in red with both length = 100 . The lag difference between each moving average is really low while VWRMA is way more precise.
Hull MA in blue and WRMA in red with both length = 100 . The Hull MA have similar overshoots than the LSMA.
Reduced overshoots moving average (ROMA) in blue and WRMA in red with both length = 100 . ROMA is an indicator i have made to reduce the overshoots of a LSMA, but at the end WRMA still reduce way more the overshoots while being smoother and having similar lag.
I have added a smoother version, just activate the extra smooth option in the indicator settings window. Here the result with length = 200 :
This result is a little bit similar to a 2 order Butterworth filter. Our filter have more overshoots which in this case could be useful to reduce the error with edges since other low pass filters tend to smooth their amplitude thus reducing edge estimation precision.
Conclusions
I have presented a well rounded filter in term of smoothness/stability and reactivity. Try to add more terms to have different results, you could maybe end up with interesting results, if its the case share them with the community :)
As for control theory i have seen neural networks integrated to Kalman flters which leaded to great accuracy, AI is everywhere and promise to be a game a changer in real time data smoothing. So i asked myself if it was possible for a neural networks to develop pinescript indicators, if yes then i could be replaced by AI ? Brrr how frightening.
Thanks for reading :)
Multi Poles Zero-Lag Exponential Moving AverageIntroduction
Based on the exponential averaging method with lag reduction, this filter allow for smoother results thanks to a multi-poles approach. Translated and modified from the Non-Linear Kalman Filter from Mladen Rakic 01/07/19 www.mql5.com
The Indicator
length control the amount of smoothing, the poles can be from 1 to 3, higher values create smoother results.
Difference With Classic Exponential Smoothing
A classic 1 depth recursion (Single smoothing) exponential moving average is defined as y = αx + (1 - α)y which can be derived into y = y + α(x - y )
2 depth recursion (Double smoothing) exponential moving average sum y with b in order to reduce the error with x , this method is calculated as follow :
y = αx + (1 - α)(y + b)
b = β(y - y ) + (1-β)b
The initial value for y is x while its 0 for b with α generally equal to 2/(length + 1)
The filter use a different approach, from the estimation of α/β/γ to the filter construction.The formula is similar to the one used in the double exponential smoothing method with a difference in y and b
y = αx + (1 - α)y
d = x - y
b = (1-β)b + d
output = y + b
instead of updating y with b the two components are directly added in a separated variable. Poles help the transition band of the frequency response to get closer to the cutoff point, the cutoff of an exponential moving average is defined as :
Cf = F/2π acos(1 - α*α/(2(1 - α)))
Also in order to minimize the overshoot of the filter a correction has been added to the output now being output = y + 1/poles * b
While this information is far being helpful to you it simply say that poles help you filter a great amount of noise thus removing irregularities of the filter.
Conclusion
The filter is interesting and while being similar to multi-depth recursion smoothing allow for more varied results thanks to its 3 poles.
Feel free to send suggestions :)
Thanks for reading
Kalman SmootherA derivation of the Kalman Filter.
Lower Gain values create smoother results.The ratio Smoothing/Lag is similar to any Low Lagging Filters.
The Gain parameter can be decimal numbers.
Kalman Smoothing With Gain = 20
For any questions/suggestions feel free to contact me
One Dimensional Parametric Kalman FilterA One Dimensional Kalman Filter, the particularity of Kalman Filtering is the constant recalculation of the Error between the measurements and the estimate.This version is modified to allow more/less filtering using an alternative calculation of the error measurement.
Camparison of the Kalman filter Red with a moving average Black of both period 50
Can be used as source for others indicators such as stochastic/rsi/moving averages...etc
For any questions/suggestions feel free to contact me
