Fourier Extrapolator of Variety RSI w/ Bollinger Bands [Loxx]Fourier Extrapolator of Variety RSI w/ Bollinger Bands is an RSI indicator that shows the original RSI, the Fourier Extrapolation of RSI in the past, and then the projection of the Fourier Extrapolated RSI for the future. This indicator has 8 different types of RSI including a new type of RSI called T3 RSI. The purpose of this indicator is to demonstrate the Fourier Extrapolation method used to model past data and to predict future price movements. This indicator will repaint. If you wish to use this for trading, then make sure to take a screenshot of the indicator when you enter the trade to save your analysis. This is the first of a series of forecasting indicators that can be used in trading. Due to how this indicator draws on the screen, you must choose values of npast and nfut that are equal to or less than 200. this is due to restrictions by TradingView and Pine Script in only allowing 500 lines on the screen at a time. Enjoy!
What is Fourier Extrapolation?
This indicator uses a multi-harmonic (or multi-tone) trigonometric model of a price series xi, i=1..n, is given by:
xi = m + Sum( a*Cos(w*i) + b*Sin(w*i), h=1..H )
Where:
xi - past price at i-th bar, total n past prices;
m - bias;
a and b - scaling coefficients of harmonics;
w - frequency of a harmonic ;
h - harmonic number;
H - total number of fitted harmonics.
Fitting this model means finding m, a, b, and w that make the modeled values to be close to real values. Finding the harmonic frequencies w is the most difficult part of fitting a trigonometric model. In the case of a Fourier series, these frequencies are set at 2*pi*h/n. But, the Fourier series extrapolation means simply repeating the n past prices into the future.
This indicator uses the Quinn-Fernandes algorithm to find the harmonic frequencies. It fits harmonics of the trigonometric series one by one until the specified total number of harmonics H is reached. After fitting a new harmonic , the coded algorithm computes the residue between the updated model and the real values and fits a new harmonic to the residue.
see here: A Fast Efficient Technique for the Estimation of Frequency , B. G. Quinn and J. M. Fernandes, Biometrika, Vol. 78, No. 3 (Sep., 1991), pp . 489-497 (9 pages) Published By: Oxford University Press
The indicator has the following input parameters:
src - input source
npast - number of past bars, to which trigonometric series is fitted;
Nfut - number of predicted future bars;
nharm - total number of harmonics in model;
frqtol - tolerance of frequency calculations.
Included:
Loxx's Expanded Source Types
Loxx's Variety RSI
Other indicators using this same method
Fourier Extrapolator of Price w/ Projection Forecast
Fourier Extrapolator of Price
Fouriertransform
Fourier Extrapolator of Price w/ Projection Forecast [Loxx]Due to popular demand, I'm pusblishing Fourier Extrapolator of Price w/ Projection Forecast.. As stated in it's twin indicator, this one is also multi-harmonic (or multi-tone) trigonometric model of a price series xi, i=1..n, is given by:
xi = m + Sum( a*Cos(w*i) + b*Sin(w*i), h=1..H )
Where:
xi - past price at i-th bar, total n past prices;
m - bias;
a and b - scaling coefficients of harmonics;
w - frequency of a harmonic ;
h - harmonic number;
H - total number of fitted harmonics.
Fitting this model means finding m, a, b, and w that make the modeled values to be close to real values. Finding the harmonic frequencies w is the most difficult part of fitting a trigonometric model. In the case of a Fourier series, these frequencies are set at 2*pi*h/n. But, the Fourier series extrapolation means simply repeating the n past prices into the future.
This indicator uses the Quinn-Fernandes algorithm to find the harmonic frequencies. It fits harmonics of the trigonometric series one by one until the specified total number of harmonics H is reached. After fitting a new harmonic , the coded algorithm computes the residue between the updated model and the real values and fits a new harmonic to the residue.
see here: A Fast Efficient Technique for the Estimation of Frequency , B. G. Quinn and J. M. Fernandes, Biometrika, Vol. 78, No. 3 (Sep., 1991), pp . 489-497 (9 pages) Published By: Oxford University Press
The indicator has the following input parameters:
src - input source
npast - number of past bars, to which trigonometric series is fitted;
Nfut - number of predicted future bars;
nharm - total number of harmonics in model;
frqtol - tolerance of frequency calculations.
The indicator plots two curves: the green/red curve indicates modeled past values and the yellow/fuchsia curve indicates the modeled future values.
The purpose of this indicator is to showcase the Fourier Extrapolator method to be used in future indicators.
Fourier Extrapolator of Price [Loxx]Fourier Extrapolator of Price is a multi-harmonic (or multi-tone) trigonometric model of a price series xi, i=1..n, is given by:
xi = m + Sum( a *Cos(w *i) + b *Sin(w *i), h=1..H )
Where:
xi - past price at i-th bar, total n past prices;
m - bias;
a and b - scaling coefficients of harmonics;
w - frequency of a harmonic;
h - harmonic number;
H - total number of fitted harmonics.
Fitting this model means finding m, a , b , and w that make the modeled values to be close to real values. Finding the harmonic frequencies w is the most difficult part of fitting a trigonometric model. In the case of a Fourier series, these frequencies are set at 2*pi*h/n. But, the Fourier series extrapolation means simply repeating the n past prices into the future.
This indicator uses the Quinn-Fernandes algorithm to find the harmonic frequencies. It fits harmonics of the trigonometric series one by one until the specified total number of harmonics H is reached. After fitting a new harmonic, the coded algorithm computes the residue between the updated model and the real values and fits a new harmonic to the residue.
see here: A Fast Efficient Technique for the Estimation of Frequency , B. G. Quinn and J. M. Fernandes, Biometrika, Vol. 78, No. 3 (Sep., 1991), pp. 489-497 (9 pages) Published By: Oxford University Press
The indicator has the following input parameters:
src - input source
npast - number of past bars, to which trigonometric series is fitted;
nharm - total number of harmonics in model;
frqtol - tolerance of frequency calculations.
The indicator plots the modeled past values
The purpose of this indicator is to showcase the Fourier Extrapolator method to be used in future indicators. While this method can also prediction future price movements, for our purpose here we will avoid doing.
JohnEhlersFourierTransformLibrary "JohnEhlersFourierTransform"
Fourier Transform for Traders By John Ehlers, slightly modified to allow to inspect other than the 8-50 frequency spectrum.
reference:
www.mesasoftware.com
high_pass_filter(source) Detrended version of the data by High Pass Filtering with a 40 Period cutoff
Parameters:
source : float, data source.
Returns: float.
transformed_dft(source, start_frequency, end_frequency) DFT by John Elhers.
Parameters:
source : float, data source.
start_frequency : int, lower bound of the frequency window, must be a positive number >= 0, window must be less than or 30.
end_frequency : int, upper bound of the frequency window, must be a positive number >= 0, window must be less than or 30.
Returns: tuple with float, float array.
db_to_rgb(db, transparency) converts the frequency decibels to rgb.
Parameters:
db : float, decibels value.
transparency : float, transparency value.
Returns: color.
FFTLibraryLibrary "FFTLibrary" contains a function for performing Fast Fourier Transform (FFT) along with a few helper functions. In general, FFT is defined for complex inputs and outputs. The real and imaginary parts of formally complex data are treated as separate arrays (denoted as x and y). For real-valued data, the array of imaginary parts should be filled with zeros.
FFT function
fft(x, y, dir) : Computes the one-dimensional discrete Fourier transform using an in-place complex-to-complex FFT algorithm . Note: The transform also produces a mirror copy of the frequency components, which correspond to the signal's negative frequencies.
Parameters:
x : float array, real part of the data, array size must be a power of 2
y : float array, imaginary part of the data, array size must be the same as x ; for real-valued input, y must be an array of zeros
dir : string, options = , defines the direction of the transform: forward" (time-to-frequency) or inverse (frequency-to-time)
Returns: x, y : tuple (float array, float array), real and imaginary parts of the transformed data (original x and y are changed on output)
Helper functions
fftPower(x, y) : Helper function that computes the power of each frequency component (in other words, Fourier amplitudes squared).
Parameters:
x : float array, real part of the Fourier amplitudes
y : float array, imaginary part of the Fourier amplitudes
Returns: power : float array of the same length as x and y , Fourier amplitudes squared
fftFreq(N) : Helper function that returns the FFT sample frequencies defined in cycles per timeframe unit. For example, if the timeframe is 5m, the frequencies are in cycles/(5 minutes).
Parameters:
N : int, window length (number of points in the transformed dataset)
Returns: freq : float array of N, contains the sample frequencies (with zero at the start).
Fast Fourier Transform (FFT) FilterDear friends!
I'm happy to present an implementation of the Fast Fourier Transform (FFT) algorithm. The script uses the FFT procedure to decompose the input time series into its cyclical constituents, in other words, its frequency components , and convert it back to the time domain with modified frequency content, that is, to filter it.
Input Description and Usage
Source and Length :
Indicates where the data comes from and the size of the lookback window used to build the dataset.
Standardize Input Dataset :
If enabled, the dataset is preprocessed by subtracting its mean and normalizing the result by the standard deviation, which is sometimes useful when analyzing seasonalities. This procedure is not recommended when using the FFT filter for smoothing (see below), as it will not preserve the average of the dataset.
Show Frequency-Domain Power Spectrum :
When enabled, the results of Fourier analysis (for the last price bar!) are plotted as a frequency-domain power spectrum , where “power” is a measure of the significance of the component in the dataset. In the spectrum, lower frequencies (longer cycles) are on the right, higher frequencies are on the left. The graph does not display the 0th component, which contains only information about the mean value. Frequency components that are allowed to pass through the filter (see below) are highlighted in magenta .
Dominant Cycles, Rows :
If this option is activated, the periods and relative powers of several dominant cyclical components that is, those that have a higher power, are listed in the table. The number of the component in the power spectrum (N) is shown in the first column. The number of rows in the table is defined by the user.
Show Inverse Fourier Transform (Filtered) :
When enabled, the reconstructed and filtered time-domain dataset (for the last price bar!) is displayed.
Apply FFT Filter in a Moving Window :
When enabled, the FFT filter with the same parameters is applied to each bar. The last data point of the reconstructed and filtered dataset is used to build a new time series. For example, by getting rid of high-frequency noise, the FFT filter can make the data smoother. By removing slowly evolving low-frequency components (including non-periodic constituents), one can reveal and analyze shorter cycles. Since filtering is done in real-time in a moving window (similar to the moving average), the modified data can potentially be used as part of a strategy and be subjected to other technical indicators.
Lowest Allowed N :
Indicates the number of the lowest frequency component used in the reconstructed time series.
Highest Allowed N :
Indicates the number of the highest frequency component used in the reconstructed time series.
Filtering Time Range block:
Specifies the time range over which real-time FFT filtering is applied. The reason for the presence of this block is that the FFT procedure is relatively computationally intensive. Therefore, the script execution may encounter the time limit imposed by TradingView when all historical bars are processed.
As always, I look forward to your feedback!
Also, leave a comment if you'd be interested in the tutorial on how to use this tool and/or in seeing the FFT filter in a strategy.
Fourier Analysis and Filtering [tbiktag]This tool uses Fourier transform to decompose the input time series into its periodic constituents and seasonalities , in other words, its frequency components . It also can reconstruct the time-domain data while using only the frequency components within a user-defined range (band-pass filtering). Thereby, this tool can reveal the cyclical characteristics of the studied market.
USAGE
The source and the size of the input data can be chosen by the “ Dataset Source ” and “ Dataset Size ” options. Price, volume, or some technical indicator (e.g., RSI, MACD, etc.) can serve as a source of the input data.
“ Action ” defines the type of the plot that will be displayed. Two options are available:
- Fourier Analysis
If selected, the frequency spectrum of the squares of the Fourier coefficient magnitudes is displayed. The zero-frequency component is on the right. Since the magnitudes of half of the coefficients are repeated, the graph displays only half of the frequency components.
The squared magnitude of a given frequency component is a measure of its power , that is, its contribution to the total variance of the dataset. Thus, by analyzing the frequency-domain spectrum, one can identify the most prominent seasonalities and then visualize them by using the " Band-pass Filtering " option (see below). Note that the zero component stores information about the amount of data, so it is naturally higher when the data is not centered at zero.
By activating the " Info about Frequency Component " option, the user can display information about the power and frequency of the selected Fourier component.
- Band-pass Filter
This option reconstructs and plots the dataset in the time domain, blocking frequency components outside of the cutoff frequencies (defined by the input parameters “ Upper Cutoff ” and “ Lower Cutoff ” input parameters in the “ Band-pass Filter Properties ” section).
FURTHER READING
In general, Fourier analysis has a long history of attempted applications for analyzing price data and estimating market cycles. For example, see the paper by John Ehlers
www.mesasoftware.com
and also some tools available here on TradingView, such as:
“Function: Discrete Fourier Transform” by @RicardoSantos
“Fourier series Model Of The Market” by @e2e4mfck
“Ehlers Discrete Fourier Transform” by @cheatcountry
Thus, I tried to make this tool versatile and user-friendly so you all can experiment with your own analysis.
Enjoy and don't hesitate to leave your feedback in the comments below!
Function: Discrete Fourier TransformExperimental:
function for inverse and discrete fourier transform in one, if you notice errors please let me know! use at your own risk...
Ehlers Discrete Fourier TransformThe Discrete Fourier Transform Indicator was written by John Ehlers and more details can be found at www.mesasoftware.com
I have color coded everything as follows: blue line is the dominant cycle, orange line is the power converted to decibels, and I have marked the other line as red if you should sell or green if you should buy
Let me know if you would like to see me write any other scripts!
FUNCTION: Goertzel algorithm -- DFT of a specific frequency binThis function implements the Goertzel algorithm (for integer N).
The Goertzel algorithm is a technique in digital signal processing (DSP) for efficient evaluation of the individual terms of the discrete Fourier transform (DFT).
In short, it measure the power of a specific frequency like one bin of a DFT, over a rolling window (N) of samples.
Here you see an input signal that changes frequency and amplitude (from 7 bars to 17). I am running the indicator 3 times to show it measuring both frequencies and one in between (13). You can see it very accurately measures the signals present and their power, but is noisy in the transition. Changing the block len will cause it to be more responsive but noisier.
Here is a picture of the same signal, but with white noise added.
If you have a cycle you think is present you could use this to test it, but the function is designed for integration in to more complicated scripts. I think power is best interrupted on a log scale.
Given a period (in bars or samples) and a block_len (N in Goertzel terminology) the function returns the Real (InPhase) and Quadrature (Imaginary) components of your signal as well as calculating the power and the instantaneous angle (in radians).
I hope this proves useful to the DSP folks here.
Low Frequency Fourier TransformThis Study uses the Real Discrete Fourier Transform algorithm to generate 3 sinusoids possibly indicative of future price.
I got information about this RDFT algorithm from "The Scientist and Engineer's Guide to Digital Signal Processing" By Steven W. Smith, Ph.D.
It has not been tested thoroughly yet, but it seems that that the RDFT isn't suited for predicting prices as the Frequency Domain Representation shows that the signal is similar to white noise, showing no significant peaks, indicative of very low periodicity of price movements.
