1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
use std::fmt;

use crate::errors::*;
use crate::{Close, Next, Reset, Update};
use std::f64::INFINITY;

/// An exponential moving average (EMA), also known as an exponentially weighted moving average
/// (EWMA).
///
/// It is a type of infinite impulse response filter that applies weighting factors which decrease exponentially.
/// The weighting for each older datum decreases exponentially, never reaching zero.
///
/// # Formula
///
/// ![EMA formula](https://wikimedia.org/api/rest_v1/media/math/render/svg/05d06bdbee2c14031fd91ead6f5f772aec1ec964)
///
/// Where:
///
/// * _EMA<sub>t</sub>_ - is the value of the EMA at any time period _t_.
/// * _EMA<sub>t-1</sub>_ - is the value of the EMA at the previous period _t-1_.
/// * _p<sub>t</sub>_ - is the input value at a time period t.
/// * _α_ - is the coefficient that represents the degree of weighting decrease, a constant smoothing factor between 0 and 1.
///
/// _α_ is calculated with the following formula:
///
/// ![alpha formula](https://wikimedia.org/api/rest_v1/media/math/render/svg/d9f6258e152db0644af548972bd6c50a8becf7ee)
///
/// Where:
///
/// * _length_ - number of periods
///
/// # Parameters
///
/// * _length_ - number of periods (integer greater than 0)
///
/// # Example
///
/// ```
/// use quantaxis_rs::indicators::ExponentialMovingAverage;
/// use quantaxis_rs::Next;
///
/// let mut ema = ExponentialMovingAverage::new(3).unwrap();
/// assert_eq!(ema.next(2.0), 2.0);
/// assert_eq!(ema.next(5.0), 3.5);
/// assert_eq!(ema.next(1.0), 2.25);
/// assert_eq!(ema.next(6.25), 4.25);
/// ```
///
/// # Links
///
/// * [Exponential moving average, Wikipedia](https://en.wikipedia.org/wiki/Moving_average#Exponential_moving_average)
///

#[derive(Debug, Clone)]
pub struct ExponentialMovingAverage {
    length: u32,
    k: f64,
    current: f64,
    is_new: bool,
    pub cached: Vec<f64>
}

impl ExponentialMovingAverage {
    pub fn new(length: u32) -> Result<Self> {
        match length {
            0 => Err(Error::from_kind(ErrorKind::InvalidParameter)),
            _ => {
                let k = 2f64 / (length as f64 + 1f64);
                let indicator = Self {
                    length,
                    k,
                    current: 0f64,
                    is_new: true,
                    cached: vec![-INFINITY; length as usize]
                };
                Ok(indicator)
            }
        }
    }

    pub fn length(&self) -> u32 {
        self.length
    }
}

impl Next<f64> for ExponentialMovingAverage {
    type Output = f64;

    fn next(&mut self, input: f64) -> Self::Output {

        if self.is_new {
            self.is_new = false;
            self.current = input;
        } else {
            self.current = self.k * input + (1.0 - self.k) * self.current;
        }

        self.cached.push(self.current.clone());
        self.cached.remove(0);
        self.current
    }

}

impl Update<f64> for ExponentialMovingAverage {
    type Output = f64;

    fn update(&mut self, input: f64) -> Self::Output {
        if self.is_new {
            self.is_new = false;
            self.current = input;
        } else {
            self.current = self.k * input + (1.0 - self.k) * self.cached[(self.length -2) as usize];
        }
        self.cached.remove((self.length - 1) as usize);
        self.cached.push(self.current.clone());
        self.current
    }

}

impl<'a, T: Close> Next<&'a T> for ExponentialMovingAverage {
    type Output = f64;

    fn next(&mut self, input: &'a T) -> Self::Output {
        self.next(input.close())
    }

}


impl<'a, T: Close> Update<&'a T> for ExponentialMovingAverage {
    type Output = f64;

    fn update(&mut self, input: &'a T) -> Self::Output {
        self.update(input.close())
    }

}

impl Reset for ExponentialMovingAverage {
    fn reset(&mut self) {
        self.current = 0.0;
        self.is_new = true;
    }
}

impl Default for ExponentialMovingAverage {
    fn default() -> Self {
        Self::new(9).unwrap()
    }
}

impl fmt::Display for ExponentialMovingAverage {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        write!(f, "EMA({})", self.length)
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::test_helper::*;
    macro_rules! test_indicator {
        ($i:tt) => {
            #[test]
            fn test_indicator() {
                let bar = Bar::new();

                // ensure Default trait is implemented
                let mut indicator = $i::default();

                // ensure Next<f64> is implemented
                let first_output = indicator.next(12.3);

                // ensure next accepts &DataItem as well
                indicator.next(&bar);

                // ensure Reset is implemented and works correctly
                indicator.reset();
                assert_eq!(indicator.next(12.3), first_output);

                // ensure Display is implemented
                format!("{}", indicator);
            }
        };
    }

    test_indicator!(ExponentialMovingAverage);

    #[test]
    fn test_new() {
        assert!(ExponentialMovingAverage::new(0).is_err());
        assert!(ExponentialMovingAverage::new(1).is_ok());
    }

    #[test]
    fn test_next() {
        let mut ema = ExponentialMovingAverage::new(2).unwrap();

        assert_eq!(ema.next(2.0), 2.0);
        assert_eq!(ema.next(5.0), 4.0);
        // assert_eq!(ema.next(1.0), 2.25);
        // assert_eq!(ema.next(6.25), 4.875);

        let mut ema = ExponentialMovingAverage::new(3).unwrap();
        let bar1 = Bar::new().close(2);
        let bar2 = Bar::new().close(5);
        assert_eq!(ema.next(&bar1), 2.0);
        assert_eq!(ema.next(&bar2), 3.5);
    }

    #[test]
    fn test_update() {
        let mut ema = ExponentialMovingAverage::new(3).unwrap();

        assert_eq!(ema.next(2.0), 2.0);
        assert_eq!(ema.next(5.0), 3.5);
        assert_eq!(ema.next(1.0), 2.25);
        assert_eq!(ema.update(6.25), 4.875);

        // let mut ema = ExponentialMovingAverage::new(3).unwrap();
        // let bar1 = Bar::new().close(2);
        // let bar2 = Bar::new().close(5);
        // assert_eq!(ema.next(&bar1), 2.0);
        // assert_eq!(ema.next(&bar2), 3.5);
    }



    #[test]
    fn test_reset() {
        let mut ema = ExponentialMovingAverage::new(5).unwrap();

        assert_eq!(ema.next(4.0), 4.0);
        ema.next(10.0);
        ema.next(15.0);
        ema.next(20.0);
        assert_ne!(ema.next(4.0), 4.0);

        ema.reset();
        assert_eq!(ema.next(4.0), 4.0);
    }

    #[test]
    fn test_default() {
        ExponentialMovingAverage::default();
    }

    #[test]
    fn test_display() {
        let ema = ExponentialMovingAverage::new(7).unwrap();
        assert_eq!(format!("{}", ema), "EMA(7)");
    }
}