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

extern crate rand;

pub mod event;

use rand::prelude::*;
use rand::distributions::Poisson;

use event::Event;

pub fn poisson_process(tmax: f64, lambda: f64) -> Vec<Event> {
    let mut rng = thread_rng();
    
    assert!(lambda >= 0.0);

    let num_events = Poisson::new(tmax*lambda).sample(&mut rng);

    let mut result = vec![];
    for _ in 0..num_events {
        let timestamp= tmax*random::<f64>();
        result.push(Event::new(timestamp));
    };
    result
}

/// Simulate a variable Poisson process
pub fn variable_poisson(tmax: f64, lambda: fn(f64) -> f64, max_lambda: f64) -> Vec<Event> {
    let mut rng = thread_rng();

    // Number of events before thinning
    let num_events = Poisson::new(tmax*max_lambda).sample(&mut rng);

    let mut result = vec![];

    // Get timestamp and intensity values of events distributed
    // according to a homogeneous Poisson process
    // and keep those who are under the intensity curve
    for _ in 0..num_events {
        let timestamp = random::<f64>()*tmax;
        let lambda_val = random::<f64>()*max_lambda;

        if lambda_val < lambda(timestamp) {
            let mut event = Event::new(timestamp);
            event.add_intensity(lambda_val);
            result.push(event);
        }
    }

    result
}

/// Simulate a Hawkes process with an exponential kernel
/// by utilising Ogata's algorithm
pub fn hawkes_exponential(tmax: f64, alpha: f64, beta: f64, lambda0: f64) -> Vec<Event> {

    let mut t = 0.0;
    let mut previous_t: f64;
    let mut last_lambda = lambda0;
    let mut result: Vec<Event> = vec![];

    while t < tmax {
        let u0 = random::<f64>();
        let s0 = -1.0/lambda0*u0.ln();

        let d = if last_lambda > lambda0 {
            1.0 + beta*u0.ln()/(last_lambda - lambda0)
        } else { std::f64::NEG_INFINITY };
        
        let s1: f64;
        
        previous_t = t;

        
        t += if d > 0.0 {
            s1 = -1.0/lambda0*random::<f64>().ln();
            if s0 <= s1 {
                s0
            } else {
                s1
            }
        } else {
            s0
        };

        if t > tmax {
            break;
        }

        last_lambda = lambda0 + alpha + (last_lambda - lambda0)*(-beta*(t-previous_t)).exp();

        let mut new_event = Event::new(t);
        new_event.add_intensity(last_lambda);

        result.push(new_event);
    }


    result
}