Crate diffusionx

DiffusionX is a multi-threaded high-performance Rust library for random number/stochastic process simulation.

Usage

use diffusionx::simulation::{Bm, Simulation, Functional};
// Brownian motion simulation
let bm = Bm::new(0.0, 1.0, 1.0)?;  // Create Brownian motion object: initial position 0, diffusion coefficient 1, duration 1
let time_step = 0.01;  // Time step
let (times, positions) = bm.simulate(time_step)?;  // Simulate Brownian motion trajectory
// Monte Carlo simulation of Brownian motion statistics
let mean = bm.mean(time_step, 1000)?;  // Mean  bm.raw_moment(time_step, 1, 1000)?;
let msd = bm.msd(time_step, 1000)?;  // Mean square displacement  bm.central_moment(time_step, 2, 1000)?;
// First passage time of Brownian motion
let max_duration = 1000; // if over this duration, the simulation will be terminated and return None
let fpt = bm.fpt(time_step, (-1.0, 1.0), max_duration)?;

Progress

Random Number Generation

• Normal distribution
• Uniform distribution
• Exponential distribution
• Poisson distribution
• Alpha-stable distribution

Stochastic Processes

• Brownian motion
• Alpha-stable Lévy process
• Subordinator
• Inverse Subordinator
• Fractional Brownian motion
• Poisson process
• Compound Poisson process
• Langevin equation
• Generalized Langevin equation
• Subordinated Langevin equation

Functional

• First passage time
• Occupation time

License

This project is dual-licensed under:

• MIT License
• Apache License Version 2.0

You can choose to use either license.

Modules

random

Random number generation

simulation

Simulation module Now implemented:

utils

Enums

SimulationError

Error for simulating the process

StableError

Error for sampling from the stable distribution

XError

Error type for the crate

Type Aliases

XResult

Result type for the crate