quantwave_core/indicators/
fisher_high_pass.rs

1use crate::indicators::high_pass::HighPass;
2use crate::indicators::metadata::{IndicatorMetadata, ParamDef};
3use crate::traits::Next;
4use crate::utils::RingBuffer as VecDeque;
5
6/// Fisher HighPass Indicator
7///
8/// Based on John Ehlers' "Inferring Trading Strategies from Probability Distribution Functions".
9/// Applies a HighPass filter, normalizes the result to [-1, 1], smooths it with a 3-tap FIR,
10/// and then applies the Fisher Transform.
11#[derive(Debug, Clone)]
12pub struct FisherHighPass {
13    hp: HighPass,
14    period: usize,
15    hp_window: VecDeque<f64>,
16    smooth_history: [f64; 2],
17    count: usize,
18}
19
20impl FisherHighPass {
21    pub fn new(hp_len: usize, norm_len: usize) -> Self {
22        Self {
23            hp: HighPass::new(hp_len),
24            period: norm_len,
25            hp_window: VecDeque::with_capacity(norm_len),
26            smooth_history: [0.0; 2],
27            count: 0,
28        }
29    }
30}
31
32impl Default for FisherHighPass {
33    fn default() -> Self {
34        Self::new(20, 20)
35    }
36}
37
38impl Next<f64> for FisherHighPass {
39    type Output = f64;
40
41    fn next(&mut self, input: f64) -> Self::Output {
42        self.count += 1;
43        let hp_val = self.hp.next(input);
44
45        self.hp_window.push_front(hp_val);
46        if self.hp_window.len() > self.period {
47            self.hp_window.pop_back();
48        }
49
50        if self.hp_window.len() < self.period {
51            return 0.0;
52        }
53
54        let mut high = f64::MIN;
55        let mut low = f64::MAX;
56        for &v in &self.hp_window {
57            if v > high {
58                high = v;
59            }
60            if v < low {
61                low = v;
62            }
63        }
64
65        let normalized = if high != low {
66            2.0 * (hp_val - low) / (high - low) - 1.0
67        } else {
68            0.0
69        };
70
71        // 3-tap FIR smoothing: (N + N[1] + N[2]) / 3
72        let smoothed = (normalized + self.smooth_history[0] + self.smooth_history[1]) / 3.0;
73
74        self.smooth_history[1] = self.smooth_history[0];
75        self.smooth_history[0] = normalized;
76
77        // Fisher Transform
78        // y = 0.5 * ln((1+x)/(1-x))
79        // Clip to avoid log(0)
80        let x = smoothed.clamp(-0.999, 0.999);
81        0.5 * ((1.0 + x) / (1.0 - x)).ln()
82    }
83}
84
85pub const FISHER_HIGH_PASS_METADATA: IndicatorMetadata = IndicatorMetadata {
86    name: "FisherHighPass",
87    description: "Fisher Transform applied to normalized HighPass filtered prices.",
88    usage: "Use to isolate high-frequency momentum from the cyclical component of price after trend removal. Provides a purer momentum signal than standard Fisher Transform applied to raw price.",
89    keywords: &["oscillator", "ehlers", "dsp", "high-pass", "momentum"],
90    ehlers_summary: "FisherHighPass applies the Fisher Transform to the high-pass filtered price rather than raw price. By first removing the low-frequency trend component with a high-pass filter, the resulting Fisher output captures only the cycle-domain momentum, producing an oscillator that is unaffected by the prevailing trend direction.",
91    params: &[
92        ParamDef {
93            name: "hp_len",
94            default: "20",
95            description: "HighPass filter length",
96        },
97        ParamDef {
98            name: "norm_len",
99            default: "20",
100            description: "Normalization lookback period",
101        },
102    ],
103    formula_source: "https://github.com/lavs9/quantwave/blob/main/references/Ehlers%20Papers/InferringTradingStrategies.pdf",
104    formula_latex: r#"
105\[
106HP = \text{HighPass}(Price, hp\_len)
107\]
108\[
109N = 2 \cdot \frac{HP - Low(HP, norm\_len)}{High(HP, norm\_len) - Low(HP, norm\_len)} - 1
110\]
111\[
112S = \frac{N + N_{t-1} + N_{t-2}}{3}
113\]
114\[
115Fisher = 0.5 \cdot \ln\left(\frac{1+S}{1-S}\right)
116\]
117"#,
118    gold_standard_file: "fisher_high_pass.json",
119    category: "Ehlers DSP",
120};
121
122#[cfg(test)]
123mod tests {
124    use super::*;
125    use crate::traits::Next;
126    use proptest::prelude::*;
127
128    #[test]
129    fn test_fisher_hp_basic() {
130        let mut fhp = FisherHighPass::new(20, 20);
131        for i in 0..100 {
132            let val = fhp.next(100.0 + (i as f64 * 0.1).sin());
133            assert!(!val.is_nan());
134        }
135    }
136
137    proptest! {
138        #[test]
139        fn test_fisher_hp_parity(
140            inputs in prop::collection::vec(1.0..100.0, 100..200),
141        ) {
142            let hp_len = 20;
143            let norm_len = 20;
144            let mut fhp = FisherHighPass::new(hp_len, norm_len);
145            let streaming_results: Vec<f64> = inputs.iter().map(|&x| fhp.next(x)).collect();
146
147            // Batch implementation
148            let mut batch_results = Vec::with_capacity(inputs.len());
149            let mut hp = HighPass::new(hp_len);
150            let hp_vals: Vec<f64> = inputs.iter().map(|&x| hp.next(x)).collect();
151
152            let mut norm_vals = Vec::new();
153            for i in 0..hp_vals.len() {
154                let start = if i >= norm_len - 1 { i + 1 - norm_len } else { 0 };
155                let window = &hp_vals[start..i + 1];
156
157                if window.len() < norm_len {
158                    batch_results.push(0.0);
159                    norm_vals.push(0.0);
160                    continue;
161                }
162
163                let mut high = f64::MIN;
164                let mut low = f64::MAX;
165                for &v in window {
166                    if v > high { high = v; }
167                    if v < low { low = v; }
168                }
169
170                let n = if high != low {
171                    2.0 * (hp_vals[i] - low) / (high - low) - 1.0
172                } else {
173                    0.0
174                };
175                norm_vals.push(n);
176
177                let s = (norm_vals[i] + (if i > 0 { norm_vals[i-1] } else { 0.0 }) + (if i > 1 { norm_vals[i-2] } else { 0.0 })) / 3.0;
178                let x = s.clamp(-0.999, 0.999);
179                batch_results.push(0.5 * ((1.0 + x) / (1.0 - x)).ln());
180            }
181
182            for (s, b) in streaming_results.iter().zip(batch_results.iter()) {
183                approx::assert_relative_eq!(s, b, epsilon = 1e-10);
184            }
185        }
186    }
187}
quantwave_core/indicators/fisher_high_pass.rs

quantwave_core/indicators/
fisher_high_pass.rs
