What is the difference between stochastic and deterministic?

A deterministic process always produces the same output from the same starting conditions — there is no randomness. A stochastic process includes randomness, so the same starting conditions can lead to many different outcomes, each with its own probability. For example, calculating the trajectory of a ball in a vacuum is deterministic, while modeling stock prices is stochastic because market fluctuations are unpredictable.

Stochastic — Definition, Formula & Examples

Stochastic describes any process, model, or system that involves randomness — meaning its future states are not fully determined but follow probability distributions. A stochastic outcome cannot be predicted exactly, only described in terms of likelihoods.

A process $\{X_t\}_{t \in T}$ is called stochastic if it is a collection of random variables defined on a probability space $(\Omega, \mathcal{F}, P)$ and indexed by a set $T$ (often representing time), where the value of $X_t$ at each index $t$ is governed by a probability distribution rather than being deterministic.

How It Works

When you model a system as stochastic, you acknowledge that its behavior includes inherent randomness. Instead of predicting a single outcome, you work with probabilities of different outcomes at each step or time point. A classic stochastic process is a random walk: at each step, you move up or down with some probability, and the path you trace is unpredictable in advance. To analyze stochastic systems, you use tools like expected value, variance, transition probabilities, and probability distributions to characterize the range and likelihood of possible outcomes.

Worked Example

Problem: A stock price currently sits at $100. Each day, it goes up by $2 with probability 0.6 or down by $2 with probability 0.4. Find the expected stock price after 3 days.

Step 1: Identify the stochastic model. Each day's change is a random variable taking value +2 with probability 0.6 or −2 with probability 0.4.

Step 2: Compute the expected change per day.

E[\Delta] = (0.6)(2) + (0.4)(-2) = 1.2 - 0.8 = 0.4

Step 3: After 3 independent days, the expected total change is 3 times the daily expected change.

E[\text{Total change}] = 3 \times 0.4 = 1.2

Step 4: Add the expected total change to the starting price.

E[\text{Price after 3 days}] = 100 + 1.2 = 101.2

Answer: The expected stock price after 3 days is $101.20. However, because the process is stochastic, the actual price could be $106, $102, $98, or $94 depending on the random outcomes each day.

Another Example

Problem: A particle starts at position 0 on a number line. Each second, it moves +1 with probability 0.5 or −1 with probability 0.5 (a symmetric random walk). What is the expected position after 10 steps, and what is the variance of its position?

Step 1: Each step has expected value zero since the walk is symmetric.

E[X_i] = (0.5)(1) + (0.5)(-1) = 0

Step 2: The expected position after 10 steps is the sum of 10 expected values.

E[S_{10}] = 10 \times 0 = 0

Step 3: Each step has variance 1, and the steps are independent, so variances add.

\text{Var}(X_i) = E[X_i^2] - (E[X_i])^2 = 1 - 0 = 1

Step 4: Find the total variance after 10 steps.

\text{Var}(S_{10}) = 10 \times 1 = 10

Answer: The expected position is 0, but the variance is 10 (standard deviation ≈ 3.16). The particle is equally likely to be positive or negative, but it is very unlikely to still be exactly at 0 — this illustrates how stochastic processes spread out over time.

Visualization

Why It Matters

Stochastic models are central to college courses in probability, statistics, and mathematical finance. Quantitative analysts on Wall Street use stochastic differential equations to price options and manage risk. In engineering and biology, stochastic processes model everything from signal noise to gene expression, making this concept essential wherever real-world randomness must be quantified.

Common Mistakes

Mistake: Treating a stochastic model's expected value as a guaranteed outcome.

Correction: The expected value is an average over many possible outcomes. Any single realization of a stochastic process can deviate significantly from the expectation. Always consider variance and the full distribution of outcomes.

Mistake: Assuming 'stochastic' and 'random' mean completely unpredictable.

Correction: Stochastic processes have well-defined probability structures. While individual outcomes are uncertain, statistical properties like means, variances, and transition probabilities are precisely specified and can be calculated.