Probability & Statistics Difficulty: Advanced

Stochastic Processes

Stochastic processes model systems evolving randomly over time or space. They are essential for finance, physics, operations research, and reinforcement learning.

Key Points

  • A stochastic process is a collection of random variables indexed by time or space.
  • Markov processes have memoryless future given the present.
  • Martingales model fair games and are central to stochastic calculus.

Formulas

Markov property
$$P(X_{t+1} | X_t, X_{t-1}, \dots) = P(X_{t+1} | X_t)$$
Martingale
$$\mathbb{E}[X_{t+1} | \mathcal{F}_t] = X_t$$
AR(1) process
$$X_t = \phi X_{t-1} + \varepsilon_t$$

Code Example

import numpy as np

T = 1000
X = np.zeros(T)
for t in range(1, T):
    X[t] = 0.9 * X[t-1] + np.random.randn()
print(X.mean(), X.var())  # var ~ 1/(1-0.9^2) = 5.26

Tags

  • random-processes
  • markov
  • martingales

References

  • Stochastic Processes
    Sheldon Ross · Wiley · source
  • Introduction to Stochastic Processes
    Gregory F. Lawler · Chapman and Hall/CRC · source

Knowledge Graph