libitofin 0.7.0

A ground-up Rust port of QuantLib: quantitative-finance primitives for pricing, risk, and numerical methods.
Documentation 
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

//! Inverse cumulative random sequence generator.
//!
//! Port of `ql/math/randomnumbers/inversecumulativersg.hpp`: draws a uniform
//! sequence from an inner [`SequenceGenerator`] and maps each component through
//! an [`InverseCumulative`] transform, yielding a sequence of distribution
//! deviates. The path weight is copied straight from the uniform sample
//! (`inversecumulativersg.hpp:88`).
//!
//! Divergence: QuantLib's single-argument constructor default-constructs the
//! inverse cumulative (`inversecumulativersg.hpp:60`); here [`new`] always takes
//! it explicitly, since the transform types on main carry no `Default`.

use super::rngtraits::{InverseCumulative, SequenceGenerator};
use crate::methods::montecarlo::Sample;
use crate::types::Real;

/// Maps a uniform [`SequenceGenerator`] `USG` through an inverse cumulative
/// transform `IC`.
pub struct InverseCumulativeRsg<USG, IC> {
    uniform_generator: USG,
    dimension: usize,
    x: Sample<Vec<Real>>,
    inverse_cumulative: IC,
}

impl<USG: SequenceGenerator, IC: InverseCumulative> InverseCumulativeRsg<USG, IC> {
    /// A generator drawing uniforms from `uniform_generator` and mapping them
    /// through `inverse_cumulative`.
    pub fn new(uniform_generator: USG, inverse_cumulative: IC) -> Self {
        let dimension = uniform_generator.dimension();
        InverseCumulativeRsg {
            uniform_generator,
            dimension,
            x: Sample::new(vec![0.0; dimension], 1.0),
            inverse_cumulative,
        }
    }
}

impl<USG: SequenceGenerator, IC: InverseCumulative> SequenceGenerator
    for InverseCumulativeRsg<USG, IC>
{
    fn next_sequence(&mut self) -> &Sample<Vec<Real>> {
        let sample = self.uniform_generator.next_sequence();
        self.x.weight = sample.weight;
        for i in 0..self.dimension {
            self.x.value[i] = self.inverse_cumulative.evaluate(sample.value[i]);
        }
        &self.x
    }

    fn last_sequence(&self) -> &Sample<Vec<Real>> {
        &self.x
    }

    fn dimension(&self) -> usize {
        self.dimension
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::math::distributions::normal::InverseCumulativeNormal;
    use crate::math::randomnumbers::randomsequencegenerator::RandomSequenceGenerator;

    // A sequence generator returning a preset weighted sample, so a test can
    // pin the transform and weight-propagation paths against known inputs.
    struct FixedSequenceGenerator {
        sample: Sample<Vec<Real>>,
    }

    impl SequenceGenerator for FixedSequenceGenerator {
        fn next_sequence(&mut self) -> &Sample<Vec<Real>> {
            &self.sample
        }
        fn last_sequence(&self) -> &Sample<Vec<Real>> {
            &self.sample
        }
        fn dimension(&self) -> usize {
            self.sample.value.len()
        }
    }

    #[test]
    fn maps_a_known_uniform_vector_to_the_inverse_normal() {
        let uniforms = vec![0.1, 0.25, 0.5, 0.75, 0.9];
        let usg = FixedSequenceGenerator {
            sample: Sample::new(uniforms.clone(), 1.0),
        };
        let mut rsg = InverseCumulativeRsg::new(usg, InverseCumulativeNormal::standard());
        let out = rsg.next_sequence();
        for (i, &u) in uniforms.iter().enumerate() {
            let expected = InverseCumulativeNormal::standard().value(u).unwrap();
            assert!(
                (out.value[i] - expected).abs() < 1e-15,
                "component {i}: got {}, want {expected}",
                out.value[i]
            );
        }
    }

    #[test]
    fn cross_checks_the_real_mersenne_twister_chain() {
        let mut driven = InverseCumulativeRsg::new(
            RandomSequenceGenerator::with_seed(4, 42).unwrap(),
            InverseCumulativeNormal::standard(),
        );
        let mut mirror = RandomSequenceGenerator::with_seed(4, 42).unwrap();
        let gaussians = driven.next_sequence().value.clone();
        let uniforms = mirror.next_sequence().value.clone();
        for (g, u) in gaussians.iter().zip(&uniforms) {
            let expected = InverseCumulativeNormal::standard().value(*u).unwrap();
            assert!((g - expected).abs() < 1e-15, "got {g}, want {expected}");
        }
    }

    #[test]
    fn propagates_the_uniform_sample_weight() {
        let usg = FixedSequenceGenerator {
            sample: Sample::new(vec![0.3, 0.6], 0.25),
        };
        let mut rsg = InverseCumulativeRsg::new(usg, InverseCumulativeNormal::standard());
        assert_eq!(rsg.next_sequence().weight, 0.25);
    }

    #[test]
    fn reports_the_inner_dimension() {
        let rsg = InverseCumulativeRsg::new(
            RandomSequenceGenerator::with_seed(7, 1).unwrap(),
            InverseCumulativeNormal::standard(),
        );
        assert_eq!(rsg.dimension(), 7);
    }
}