nautilus_indicators/momentum/
stochastics.rs

1// -------------------------------------------------------------------------------------------------
2//  Copyright (C) 2015-2026 Nautech Systems Pty Ltd. All rights reserved.
3//  https://nautechsystems.io
4//
5//  Licensed under the GNU Lesser General Public License Version 3.0 (the "License");
6//  You may not use this file except in compliance with the License.
7//  You may obtain a copy of the License at https://www.gnu.org/licenses/lgpl-3.0.en.html
8//
9//  Unless required by applicable law or agreed to in writing, software
10//  distributed under the License is distributed on an "AS IS" BASIS,
11//  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12//  See the License for the specific language governing permissions and
13//  limitations under the License.
14// -------------------------------------------------------------------------------------------------
15
16use std::fmt::{Debug, Display};
17
18use arraydeque::{ArrayDeque, Wrapping};
19use nautilus_model::data::Bar;
20use strum::{AsRefStr, Display as StrumDisplay, EnumIter, EnumString, FromRepr};
21
22use crate::{
23    average::{MovingAverageFactory, MovingAverageType},
24    indicator::{Indicator, MovingAverage},
25};
26
27const MAX_PERIOD: usize = 1_024;
28
29/// Method for calculating %D in the Stochastics indicator.
30///
31/// The %D line is the smoothed version of %K and can provide trading signals.
32/// Two calculation methods are supported:
33///
34/// - **Ratio**: Original Nautilus method using `100 * SUM(close-LL) / SUM(HH-LL)` over `period_d`.
35///   This is range-weighted and has less lag than MA-based methods.
36/// - **MovingAverage**: Uses MA of slowed %K values, compatible with
37///   cTrader/MetaTrader/TradingView implementations.
38#[repr(C)]
39#[derive(
40    Copy,
41    Clone,
42    Debug,
43    Default,
44    Hash,
45    PartialEq,
46    Eq,
47    PartialOrd,
48    Ord,
49    AsRefStr,
50    FromRepr,
51    EnumIter,
52    EnumString,
53    StrumDisplay,
54)]
55#[strum(ascii_case_insensitive)]
56#[strum(serialize_all = "SCREAMING_SNAKE_CASE")]
57#[cfg_attr(
58    feature = "python",
59    pyo3::pyclass(
60        frozen,
61        eq,
62        eq_int,
63        hash,
64        module = "nautilus_trader.core.nautilus_pyo3.indicators"
65    )
66)]
67pub enum StochasticsDMethod {
68    /// Ratio: Nautilus original method: `100 * SUM(close-LL) / SUM(HH-LL)` over `period_d`.
69    /// This is range-weighted and has less lag than MA-based methods.
70    #[default]
71    Ratio,
72    /// MA method: `MA(slowed_k, period_d, ma_type)`.
73    /// This produces values compatible with cTrader/MetaTrader/TradingView implementations.
74    MovingAverage,
75}
76
77#[repr(C)]
78#[cfg_attr(
79    feature = "python",
80    pyo3::pyclass(module = "nautilus_trader.core.nautilus_pyo3.indicators")
81)]
82pub struct Stochastics {
83    /// The lookback period for %K calculation (highest high / lowest low).
84    pub period_k: usize,
85    /// The smoothing period for %D calculation.
86    pub period_d: usize,
87    /// The slowing period for %K smoothing (1 = no slowing (Nautilus original).
88    pub slowing: usize,
89    /// The moving average type used for slowing and MA-based %D.
90    pub ma_type: MovingAverageType,
91    /// The method for calculating %D (Ratio = Nautilus original method, MovingAverage = MA Smoothed).
92    pub d_method: StochasticsDMethod,
93    /// The current %K value (slowed if slowing > 1).
94    pub value_k: f64,
95    /// The current %D value.
96    pub value_d: f64,
97    /// Whether the indicator has received sufficient inputs to produce valid values.
98    pub initialized: bool,
99    has_inputs: bool,
100    highs: ArrayDeque<f64, MAX_PERIOD, Wrapping>,
101    lows: ArrayDeque<f64, MAX_PERIOD, Wrapping>,
102    c_sub_1: ArrayDeque<f64, MAX_PERIOD, Wrapping>,
103    h_sub_l: ArrayDeque<f64, MAX_PERIOD, Wrapping>,
104    /// Moving average for %K slowing (None when slowing == 1).
105    slowing_ma: Option<Box<dyn MovingAverage + Send + Sync>>,
106    /// Moving average for %D when d_method == MovingAverage.
107    d_ma: Option<Box<dyn MovingAverage + Send + Sync>>,
108}
109
110impl Debug for Stochastics {
111    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
112        f.debug_struct(stringify!(Stochastics))
113            .field("period_k", &self.period_k)
114            .field("period_d", &self.period_d)
115            .field("slowing", &self.slowing)
116            .field("ma_type", &self.ma_type)
117            .field("d_method", &self.d_method)
118            .field("value_k", &self.value_k)
119            .field("value_d", &self.value_d)
120            .field("initialized", &self.initialized)
121            .field("has_inputs", &self.has_inputs)
122            .field(
123                "slowing_ma",
124                &self.slowing_ma.as_ref().map(|_| "MovingAverage"),
125            )
126            .field("d_ma", &self.d_ma.as_ref().map(|_| "MovingAverage"))
127            .finish()
128    }
129}
130
131impl Display for Stochastics {
132    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
133        write!(f, "{}({},{})", self.name(), self.period_k, self.period_d,)
134    }
135}
136
137impl Indicator for Stochastics {
138    fn name(&self) -> String {
139        stringify!(Stochastics).to_string()
140    }
141
142    fn has_inputs(&self) -> bool {
143        self.has_inputs
144    }
145
146    fn initialized(&self) -> bool {
147        self.initialized
148    }
149
150    fn handle_bar(&mut self, bar: &Bar) {
151        self.update_raw((&bar.high).into(), (&bar.low).into(), (&bar.close).into());
152    }
153
154    fn reset(&mut self) {
155        self.highs.clear();
156        self.lows.clear();
157        self.c_sub_1.clear();
158        self.h_sub_l.clear();
159        self.value_k = 0.0;
160        self.value_d = 0.0;
161        self.has_inputs = false;
162        self.initialized = false;
163
164        // Reset slowing MA if present
165        if let Some(ref mut ma) = self.slowing_ma {
166            ma.reset();
167        }
168
169        // Reset %D MA if present
170        if let Some(ref mut ma) = self.d_ma {
171            ma.reset();
172        }
173    }
174}
175
176impl Stochastics {
177    /// Creates a new [`Stochastics`] instance with default parameters.
178    ///
179    /// This is the backward-compatible constructor that produces identical output
180    /// to the original Nautilus implementation, setting the following to:
181    /// - `slowing = 1` (no slowing applied to %K)
182    /// - `ma_type = Exponential` (unused when slowing = 1 or with Ratio method)
183    /// - `d_method = Ratio` (Nautilus native %D calculation)
184    ///
185    /// # Panics
186    ///
187    /// This function panics if:
188    /// - `period_k` or `period_d` is less than 1 or greater than `MAX_PERIOD`.
189    #[must_use]
190    pub fn new(period_k: usize, period_d: usize) -> Self {
191        Self::new_with_params(
192            period_k,
193            period_d,
194            1,                              // slowing = 1 (no slowing)
195            MovingAverageType::Exponential, // ma_type (unused)
196            StochasticsDMethod::Ratio,      // d_method = Ratio
197        )
198    }
199
200    /// Creates a new [`Stochastics`] instance with full parameter control.
201    ///
202    /// # Parameters
203    ///
204    /// - `period_k`: The lookback period for %K (highest high / lowest low).
205    /// - `period_d`: The smoothing period for %D.
206    /// - `slowing`: MA smoothing period for raw %K (1 = no slowing, > 1 = smoothed).
207    /// - `ma_type`: MA type for slowing and MA-based %D (EMA, SMA, Wilder, etc.).
208    /// - `d_method`: %D calculation method (Ratio = Nautilus original, MovingAverage = MA smoothed).
209    ///
210    /// # Panics
211    ///
212    /// This function panics if:
213    /// - `period_k`, `period_d`, or `slowing` is less than 1 or greater than `MAX_PERIOD`.
214    #[must_use]
215    pub fn new_with_params(
216        period_k: usize,
217        period_d: usize,
218        slowing: usize,
219        ma_type: MovingAverageType,
220        d_method: StochasticsDMethod,
221    ) -> Self {
222        assert!(
223            period_k > 0 && period_k <= MAX_PERIOD,
224            "Stochastics: period_k {period_k} exceeds bounds (1..={MAX_PERIOD})"
225        );
226        assert!(
227            period_d > 0 && period_d <= MAX_PERIOD,
228            "Stochastics: period_d {period_d} exceeds bounds (1..={MAX_PERIOD})"
229        );
230        assert!(
231            slowing > 0 && slowing <= MAX_PERIOD,
232            "Stochastics: slowing {slowing} exceeds bounds (1..={MAX_PERIOD})"
233        );
234
235        // Create slowing MA only if slowing > 1
236        let slowing_ma = if slowing > 1 {
237            Some(MovingAverageFactory::create(ma_type, slowing))
238        } else {
239            None
240        };
241
242        // Create %D MA only if d_method == MovingAverage
243        let d_ma = match d_method {
244            StochasticsDMethod::MovingAverage => {
245                Some(MovingAverageFactory::create(ma_type, period_d))
246            }
247            StochasticsDMethod::Ratio => None,
248        };
249
250        Self {
251            period_k,
252            period_d,
253            slowing,
254            ma_type,
255            d_method,
256            has_inputs: false,
257            initialized: false,
258            value_k: 0.0,
259            value_d: 0.0,
260            highs: ArrayDeque::new(),
261            lows: ArrayDeque::new(),
262            h_sub_l: ArrayDeque::new(),
263            c_sub_1: ArrayDeque::new(),
264            slowing_ma,
265            d_ma,
266        }
267    }
268
269    /// Updates the indicator with raw price values.
270    ///
271    /// # Parameters
272    ///
273    /// - `high`: The high price for the period.
274    /// - `low`: The low price for the period.
275    /// - `close`: The close price for the period.
276    pub fn update_raw(&mut self, high: f64, low: f64, close: f64) {
277        if !self.has_inputs {
278            self.has_inputs = true;
279        }
280
281        // Maintain high/low deques for period_k lookback
282        if self.highs.len() == self.period_k {
283            self.highs.pop_front();
284            self.lows.pop_front();
285        }
286        let _ = self.highs.push_back(high);
287        let _ = self.lows.push_back(low);
288
289        // Check initialization for period_k (matches original behavior)
290        if !self.initialized
291            && self.highs.len() == self.period_k
292            && self.lows.len() == self.period_k
293        {
294            // Original behavior: set initialized when period_k is filled
295            // (for backward compat with d_method=Ratio, slowing=1)
296            if self.slowing_ma.is_none() && self.d_method == StochasticsDMethod::Ratio {
297                self.initialized = true;
298            }
299        }
300
301        // Calculate highest high and lowest low over period_k
302        let k_max_high = self.highs.iter().copied().fold(f64::NEG_INFINITY, f64::max);
303        let k_min_low = self.lows.iter().copied().fold(f64::INFINITY, f64::min);
304
305        // For Ratio method, always update the deques (matches original behavior)
306        if self.d_method == StochasticsDMethod::Ratio {
307            if self.c_sub_1.len() == self.period_d {
308                self.c_sub_1.pop_front();
309                self.h_sub_l.pop_front();
310            }
311            let _ = self.c_sub_1.push_back(close - k_min_low);
312            let _ = self.h_sub_l.push_back(k_max_high - k_min_low);
313        }
314
315        // Handle division by zero (flat market)
316        if k_max_high == k_min_low {
317            return;
318        }
319
320        // Calculate raw %K
321        let raw_k = 100.0 * ((close - k_min_low) / (k_max_high - k_min_low));
322
323        // Apply slowing if configured (slowing > 1)
324        let slowed_k = match &mut self.slowing_ma {
325            Some(ma) => {
326                ma.update_raw(raw_k);
327                ma.value()
328            }
329            None => raw_k, // No slowing when slowing == 1
330        };
331        self.value_k = slowed_k;
332
333        // Calculate %D based on d_method
334        self.value_d = match self.d_method {
335            StochasticsDMethod::Ratio => {
336                // Nautilus original: 100 * SUM(close-LL) / SUM(HH-LL) over period_d
337                // Deques already updated above
338                let sum_h_sub_l: f64 = self.h_sub_l.iter().sum();
339                if sum_h_sub_l == 0.0 {
340                    0.0
341                } else {
342                    100.0 * (self.c_sub_1.iter().sum::<f64>() / sum_h_sub_l)
343                }
344            }
345            StochasticsDMethod::MovingAverage => {
346                // cTrader-like: MA(slowed_k, period_d, ma_type)
347                if let Some(ref mut ma) = self.d_ma {
348                    ma.update_raw(slowed_k);
349                    ma.value()
350                } else {
351                    50.0 // Fallback (shouldn't happen)
352                }
353            }
354        };
355
356        // Update initialization state for new parameter combinations
357        // For slowing > 1, we need additional warmup for the slowing MA
358        // For d_method == MovingAverage, we need additional warmup for the %D MA
359        if !self.initialized {
360            let base_ready = self.highs.len() == self.period_k;
361            let slowing_ready = match &self.slowing_ma {
362                Some(ma) => ma.initialized(),
363                None => true,
364            };
365            let d_ready = match self.d_method {
366                StochasticsDMethod::Ratio => true, // Already handled above for backward compat
367                StochasticsDMethod::MovingAverage => match &self.d_ma {
368                    Some(ma) => ma.initialized(),
369                    None => true,
370                },
371            };
372
373            if base_ready && slowing_ready && d_ready {
374                self.initialized = true;
375            }
376        }
377    }
378}
379
380#[cfg(test)]
381mod tests {
382    use nautilus_model::data::Bar;
383    use rstest::rstest;
384
385    use crate::{
386        average::MovingAverageType,
387        indicator::Indicator,
388        momentum::stochastics::{Stochastics, StochasticsDMethod},
389        stubs::{bar_ethusdt_binance_minute_bid, stochastics_10},
390    };
391
392    #[rstest]
393    fn test_stochastics_initialized(stochastics_10: Stochastics) {
394        let display_str = format!("{stochastics_10}");
395        assert_eq!(display_str, "Stochastics(10,10)");
396        assert_eq!(stochastics_10.period_d, 10);
397        assert_eq!(stochastics_10.period_k, 10);
398        assert!(!stochastics_10.initialized);
399        assert!(!stochastics_10.has_inputs);
400    }
401
402    #[rstest]
403    fn test_value_with_one_input(mut stochastics_10: Stochastics) {
404        stochastics_10.update_raw(1.0, 1.0, 1.0);
405        assert_eq!(stochastics_10.value_d, 0.0);
406        assert_eq!(stochastics_10.value_k, 0.0);
407    }
408
409    #[rstest]
410    fn test_value_with_three_inputs(mut stochastics_10: Stochastics) {
411        stochastics_10.update_raw(1.0, 1.0, 1.0);
412        stochastics_10.update_raw(2.0, 2.0, 2.0);
413        stochastics_10.update_raw(3.0, 3.0, 3.0);
414        assert_eq!(stochastics_10.value_d, 100.0);
415        assert_eq!(stochastics_10.value_k, 100.0);
416    }
417
418    #[rstest]
419    fn test_value_with_ten_inputs(mut stochastics_10: Stochastics) {
420        let high_values = [
421            1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0, 14.0, 15.0,
422        ];
423        let low_values = [
424            0.9, 1.9, 2.9, 3.9, 4.9, 5.9, 6.9, 7.9, 8.9, 9.9, 10.1, 10.2, 10.3, 11.1, 11.4,
425        ];
426        let close_values = [
427            1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0, 13.0, 14.0, 15.0,
428        ];
429
430        for i in 0..15 {
431            stochastics_10.update_raw(high_values[i], low_values[i], close_values[i]);
432        }
433
434        assert!(stochastics_10.initialized());
435        assert_eq!(stochastics_10.value_d, 100.0);
436        assert_eq!(stochastics_10.value_k, 100.0);
437    }
438
439    #[rstest]
440    fn test_initialized_with_required_input(mut stochastics_10: Stochastics) {
441        for i in 1..10 {
442            stochastics_10.update_raw(f64::from(i), f64::from(i), f64::from(i));
443        }
444        assert!(!stochastics_10.initialized);
445        stochastics_10.update_raw(10.0, 12.0, 14.0);
446        assert!(stochastics_10.initialized);
447    }
448
449    #[rstest]
450    fn test_handle_bar(mut stochastics_10: Stochastics, bar_ethusdt_binance_minute_bid: Bar) {
451        stochastics_10.handle_bar(&bar_ethusdt_binance_minute_bid);
452        assert_eq!(stochastics_10.value_d, 49.090_909_090_909_09);
453        assert_eq!(stochastics_10.value_k, 49.090_909_090_909_09);
454        assert!(stochastics_10.has_inputs);
455        assert!(!stochastics_10.initialized);
456    }
457
458    #[rstest]
459    fn test_reset(mut stochastics_10: Stochastics) {
460        stochastics_10.update_raw(1.0, 1.0, 1.0);
461        assert_eq!(stochastics_10.c_sub_1.len(), 1);
462        assert_eq!(stochastics_10.h_sub_l.len(), 1);
463
464        stochastics_10.reset();
465        assert_eq!(stochastics_10.value_d, 0.0);
466        assert_eq!(stochastics_10.value_k, 0.0);
467        assert_eq!(stochastics_10.h_sub_l.len(), 0);
468        assert_eq!(stochastics_10.c_sub_1.len(), 0);
469        assert!(!stochastics_10.has_inputs);
470        assert!(!stochastics_10.initialized);
471    }
472
473    #[rstest]
474    fn test_new_defaults_slowing_1_ratio() {
475        let stoch = Stochastics::new(10, 3);
476        assert_eq!(stoch.period_k, 10);
477        assert_eq!(stoch.period_d, 3);
478        assert_eq!(stoch.slowing, 1);
479        assert_eq!(stoch.ma_type, MovingAverageType::Exponential);
480        assert_eq!(stoch.d_method, StochasticsDMethod::Ratio);
481        assert!(
482            stoch.slowing_ma.is_none(),
483            "slowing_ma should be None when slowing == 1"
484        );
485        assert!(
486            stoch.d_ma.is_none(),
487            "d_ma should be None when d_method == Ratio"
488        );
489    }
490
491    #[rstest]
492    fn test_new_with_params_accepts_all_params() {
493        let stoch = Stochastics::new_with_params(
494            11,
495            3,
496            3,
497            MovingAverageType::Exponential,
498            StochasticsDMethod::MovingAverage,
499        );
500        assert_eq!(stoch.period_k, 11);
501        assert_eq!(stoch.period_d, 3);
502        assert_eq!(stoch.slowing, 3);
503        assert_eq!(stoch.ma_type, MovingAverageType::Exponential);
504        assert_eq!(stoch.d_method, StochasticsDMethod::MovingAverage);
505        assert!(
506            stoch.slowing_ma.is_some(),
507            "slowing_ma should exist when slowing > 1"
508        );
509        assert!(
510            stoch.d_ma.is_some(),
511            "d_ma should exist when d_method == MovingAverage"
512        );
513    }
514
515    #[rstest]
516    fn test_backward_compatibility_identical_output() {
517        // Create both old-style and new-style with explicit defaults
518        let mut stoch_old = Stochastics::new(10, 10);
519        let mut stoch_new = Stochastics::new_with_params(
520            10,
521            10,
522            1,
523            MovingAverageType::Exponential,
524            StochasticsDMethod::Ratio,
525        );
526
527        // Feed identical data to both
528        let high_values = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0];
529        let low_values = [0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5, 9.5];
530        let close_values = [0.8, 1.8, 2.8, 3.8, 4.8, 5.8, 6.8, 7.8, 8.8, 9.8];
531
532        for i in 0..10 {
533            stoch_old.update_raw(high_values[i], low_values[i], close_values[i]);
534            stoch_new.update_raw(high_values[i], low_values[i], close_values[i]);
535        }
536
537        // Output should be bit-for-bit identical
538        assert_eq!(stoch_old.value_k, stoch_new.value_k, "value_k mismatch");
539        assert_eq!(stoch_old.value_d, stoch_new.value_d, "value_d mismatch");
540        assert_eq!(stoch_old.initialized, stoch_new.initialized);
541    }
542
543    #[rstest]
544    fn test_slowing_3_smoothes_k() {
545        let mut stoch_no_slowing = Stochastics::new(5, 3);
546        let mut stoch_with_slowing = Stochastics::new_with_params(
547            5,
548            3,
549            3,
550            MovingAverageType::Exponential,
551            StochasticsDMethod::Ratio,
552        );
553
554        // Generate varying data to show smoothing effect
555        let data = [
556            (10.0, 5.0, 8.0),
557            (12.0, 6.0, 7.0),
558            (11.0, 4.0, 9.0),
559            (13.0, 7.0, 8.0),
560            (14.0, 8.0, 10.0),
561            (12.0, 6.0, 7.0),
562            (15.0, 9.0, 14.0),
563            (16.0, 10.0, 11.0),
564        ];
565
566        for (high, low, close) in data {
567            stoch_no_slowing.update_raw(high, low, close);
568            stoch_with_slowing.update_raw(high, low, close);
569        }
570
571        // With slowing, %K should be smoother (different from raw)
572        // We can't assert exact values without knowing the expected behavior,
573        // but we can verify they differ when slowing is applied
574        assert!(
575            (stoch_no_slowing.value_k - stoch_with_slowing.value_k).abs() > 0.01,
576            "Slowing should produce different %K values"
577        );
578    }
579
580    #[rstest]
581    #[case(MovingAverageType::Simple)]
582    #[case(MovingAverageType::Exponential)]
583    #[case(MovingAverageType::Wilder)]
584    #[case(MovingAverageType::Hull)]
585    fn test_slowing_with_different_ma_types(#[case] ma_type: MovingAverageType) {
586        let mut stoch = Stochastics::new_with_params(5, 3, 3, ma_type, StochasticsDMethod::Ratio);
587
588        // Feed data and verify it produces valid output
589        for i in 1..=10 {
590            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
591        }
592
593        assert!(
594            stoch.value_k.is_finite(),
595            "value_k should be finite with {ma_type:?}"
596        );
597        assert!(
598            stoch.value_d.is_finite(),
599            "value_d should be finite with {ma_type:?}"
600        );
601        assert!(
602            stoch.value_k >= 0.0 && stoch.value_k <= 100.0,
603            "value_k out of range with {ma_type:?}"
604        );
605    }
606
607    #[rstest]
608    fn test_d_method_ratio_preserves_nautilus_behavior() {
609        let mut stoch = Stochastics::new_with_params(
610            10,
611            3,
612            1, // No slowing
613            MovingAverageType::Exponential,
614            StochasticsDMethod::Ratio,
615        );
616
617        // Same data as original test
618        for i in 1..=15 {
619            stoch.update_raw(f64::from(i), f64::from(i) - 0.1, f64::from(i));
620        }
621
622        // Should produce same ratio-based %D as original
623        assert!(stoch.initialized);
624        assert!(stoch.value_d > 0.0);
625    }
626
627    #[rstest]
628    fn test_d_method_ma_produces_smoothed_k() {
629        let mut stoch = Stochastics::new_with_params(
630            5,
631            3,
632            3, // With slowing
633            MovingAverageType::Exponential,
634            StochasticsDMethod::MovingAverage, // MA-based %D
635        );
636
637        let data = [
638            (10.0, 5.0, 8.0),
639            (12.0, 6.0, 7.0),
640            (11.0, 4.0, 9.0),
641            (13.0, 7.0, 8.0),
642            (14.0, 8.0, 10.0),
643            (12.0, 6.0, 7.0),
644            (15.0, 9.0, 14.0),
645            (16.0, 10.0, 11.0),
646            (14.0, 8.0, 12.0),
647            (13.0, 7.0, 10.0),
648        ];
649
650        for (high, low, close) in data {
651            stoch.update_raw(high, low, close);
652        }
653
654        // %D should be smoothed version of %K
655        assert!(stoch.value_d.is_finite());
656        assert!(stoch.value_d >= 0.0 && stoch.value_d <= 100.0);
657    }
658
659    #[rstest]
660    fn test_warmup_period_with_slowing() {
661        let mut stoch = Stochastics::new_with_params(
662            5,
663            3,
664            3, // slowing = 3 means we need period_k + slowing inputs for slowing MA
665            MovingAverageType::Exponential,
666            StochasticsDMethod::Ratio,
667        );
668
669        // With period_k=5, slowing=3, period_d=3:
670        // - Need 5 bars for period_k
671        // - Need 3 more for slowing MA to initialize
672        // - Need 3 for period_d ratio
673        // Exact warmup depends on MA implementation
674
675        for i in 1..=4 {
676            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
677            assert!(!stoch.initialized, "Should not be initialized at bar {i}");
678        }
679
680        // After enough bars, should initialize
681        for i in 5..=15 {
682            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
683        }
684
685        assert!(
686            stoch.initialized,
687            "Should be initialized after sufficient bars"
688        );
689    }
690
691    #[rstest]
692    fn test_warmup_period_with_ma_d_method() {
693        let mut stoch = Stochastics::new_with_params(
694            5,
695            3,
696            3,
697            MovingAverageType::Exponential,
698            StochasticsDMethod::MovingAverage, // MA %D needs its own warmup
699        );
700
701        for i in 1..=4 {
702            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
703        }
704        assert!(!stoch.initialized);
705
706        // Keep feeding until initialized
707        for i in 5..=20 {
708            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
709        }
710
711        assert!(
712            stoch.initialized,
713            "Should be initialized after sufficient bars"
714        );
715    }
716
717    #[rstest]
718    fn test_reset_clears_slowing_ma_state() {
719        let mut stoch = Stochastics::new_with_params(
720            5,
721            3,
722            3,
723            MovingAverageType::Exponential,
724            StochasticsDMethod::MovingAverage,
725        );
726
727        // Feed some data
728        for i in 1..=10 {
729            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
730        }
731
732        assert!(stoch.has_inputs);
733
734        // Reset
735        stoch.reset();
736
737        assert!(!stoch.has_inputs);
738        assert!(!stoch.initialized);
739        assert_eq!(stoch.value_k, 0.0);
740        assert_eq!(stoch.value_d, 0.0);
741        assert_eq!(stoch.highs.len(), 0);
742        assert_eq!(stoch.lows.len(), 0);
743
744        // After reset, should be able to use again
745        for i in 1..=10 {
746            stoch.update_raw(f64::from(i) + 5.0, f64::from(i), f64::from(i) + 2.0);
747        }
748        assert!(stoch.value_k > 0.0);
749    }
750
751    #[rstest]
752    fn test_slowing_1_bypasses_ma() {
753        let stoch = Stochastics::new_with_params(
754            10,
755            3,
756            1, // slowing = 1 means no MA
757            MovingAverageType::Exponential,
758            StochasticsDMethod::Ratio,
759        );
760
761        assert!(
762            stoch.slowing_ma.is_none(),
763            "slowing = 1 should not create MA"
764        );
765    }
766
767    #[rstest]
768    #[should_panic(expected = "slowing")]
769    fn test_slowing_0_panics() {
770        let _ = Stochastics::new_with_params(
771            10,
772            3,
773            0, // Invalid
774            MovingAverageType::Exponential,
775            StochasticsDMethod::Ratio,
776        );
777    }
778
779    #[rstest]
780    fn test_division_by_zero_protection() {
781        let mut stoch = Stochastics::new_with_params(
782            5,
783            3,
784            3,
785            MovingAverageType::Exponential,
786            StochasticsDMethod::MovingAverage,
787        );
788
789        // Flat market: high == low == close
790        for _ in 0..10 {
791            stoch.update_raw(100.0, 100.0, 100.0);
792        }
793
794        // Should not panic, values should be 0 or previous
795        assert!(stoch.value_k.is_finite());
796        assert!(stoch.value_d.is_finite());
797    }
798}
nautilus_indicators/momentum/stochastics.rs

nautilus_indicators/momentum/
stochastics.rs
