Lean
$LEAN_TAG$
|
The Fisher transform is a mathematical process which is used to convert any data set to a modified data set whose Probability Distribution Function is approximately Gaussian. Once the Fisher transform is computed, the transformed data can then be analyzed in terms of it's deviation from the mean. More...
Public Member Functions | |
FisherTransform (int period) | |
Initializes a new instance of the FisherTransform class with the default name and period More... | |
FisherTransform (string name, int period) | |
A Fisher Transform of Prices More... | |
override void | Reset () |
Resets this indicator to its initial state More... | |
Public Attributes | |
override bool | IsReady => _medianMax.IsReady && _medianMax.IsReady |
Gets a flag indicating when this indicator is ready and fully initialized More... | |
Protected Member Functions | |
override decimal | ComputeNextValue (IBaseDataBar input) |
Computes the next value in the transform. value1 is a function used to normalize price withing the last _period day range. value1 is centered on its midpoint and then doubled so that value1 wil swing between -1 and +1. value1 is also smoothed with an exponential moving average whose alpha is 0.33. More... | |
Protected Member Functions inherited from QuantConnect.Indicators.BarIndicator | |
BarIndicator (string name) | |
Creates a new TradeBarIndicator with the specified name More... | |
Properties | |
int | WarmUpPeriod [get] |
Required period, in data points, for the indicator to be ready and fully initialized. More... | |
Properties inherited from QuantConnect.Indicators.IIndicatorWarmUpPeriodProvider | |
int | WarmUpPeriod [get] |
Required period, in data points, for the indicator to be ready and fully initialized. More... | |
The Fisher transform is a mathematical process which is used to convert any data set to a modified data set whose Probability Distribution Function is approximately Gaussian. Once the Fisher transform is computed, the transformed data can then be analyzed in terms of it's deviation from the mean.
The equation is y = .5 * ln [ 1 + x / 1 - x ] where x is the input y is the output ln is the natural logarithm
The Fisher transform has much sharper turning points than other indicators such as MACD
For more info, read chapter 1 of Cybernetic Analysis for Stocks and Futures by John F. Ehlers
We are implementing the latest version of this indicator found at Fig. 4 of http://www.mesasoftware.com/papers/UsingTheFisherTransform.pdf
Definition at line 39 of file FisherTransform.cs.
QuantConnect.Indicators.FisherTransform.FisherTransform | ( | int | period | ) |
Initializes a new instance of the FisherTransform class with the default name and period
period | The period of the WMA |
Definition at line 50 of file FisherTransform.cs.
QuantConnect.Indicators.FisherTransform.FisherTransform | ( | string | name, |
int | period | ||
) |
A Fisher Transform of Prices
name | string - the name of the indicator |
period | The number of periods for the indicator |
Definition at line 60 of file FisherTransform.cs.
|
protected |
Computes the next value in the transform. value1 is a function used to normalize price withing the last _period day range. value1 is centered on its midpoint and then doubled so that value1 wil swing between -1 and +1. value1 is also smoothed with an exponential moving average whose alpha is 0.33.
Since the smoothing may allow value1 to exceed the _period day price range, limits are introduced to preclude the transform from blowing up by having an input larger than unity.
input | IndicatorDataPoint - the time and value of the next price |
Definition at line 89 of file FisherTransform.cs.
override void QuantConnect.Indicators.FisherTransform.Reset | ( | ) |
Resets this indicator to its initial state
Definition at line 115 of file FisherTransform.cs.
override bool QuantConnect.Indicators.FisherTransform.IsReady => _medianMax.IsReady && _medianMax.IsReady |
Gets a flag indicating when this indicator is ready and fully initialized
Definition at line 72 of file FisherTransform.cs.
|
get |
Required period, in data points, for the indicator to be ready and fully initialized.
Definition at line 77 of file FisherTransform.cs.