How do I simulate a fraud score?

Open this page and use the live Time-Series Forecaster tool below — set your inputs and the simulation runs instantly in your browser using real numerics. No install, no account needed.

Yes. The simulation runs free in your browser. A one-time unlock or a Pro plan adds advanced parameters, saved presets, data import, and clean exports.

Can I use my own numbers?

Absolutely — every input is adjustable, and with data import you can drive a fraud score from your own measurements.

Use case · powered by Time-Series Forecaster

Time-Series Forecaster for a fraud score

Simulate a fraud score live in your browser. This runs the real Time-Series Forecaster solver — adjust the inputs, watch it respond instantly, and export the result. No install, no account.

For:Students Researchers Educators Engineers Hobbyists & Makers K-12 Students First Responders

Linear Regression

open full solver →

Linear RegressionLive

least squares & R²

Controls

Presets

True slope

Noise level

Data points

Ordinary least squares finds the line that minimizes the sum of squared vertical residuals (the gray drop-lines). R² measures the fraction of variance the line explains — 1 is perfect, 0 is useless. Add noise or remove points to watch the fit and R² degrade.

▶ Run in Python

Data Inspector

Fitted slope0.000

Intercept0.000

R²0.000

Correlation r0.000

Governing equation

\overset{y}{^} = 0.00 + 0.00 x, β_{1} = \frac{\sum ( x - x ˉ ) ( y - y ˉ )}{\sum ( x - x ˉ ) ^{2}}, R^{2} = 0.00

Reading this result: R² = 0.00: noise (2) swamps the signal, so least squares can barely tell the slope from flat — adding data points would tighten the estimate.

Runs locally in your browser — free forever. Scale to the cloud when reality gets heavy.

↓

Fourier

open full solver →

Fourier Series BuilderLive

synthesize waves from sinusoids

Controls

Add harmonics and watch sine waves sum into a square, sawtooth, or triangle wave — the Fourier series in action.

Presets

Harmonics

▶ Run in Python

Data Inspector

Wavesquare

Harmonics8

Basissine

Governing equation

f (t) = \frac{a _{0}}{2} + n = 1 \sum 8 (a_{n} cos nω t + b_{n} sin nω t)

Reading this result: Reconstructing the square wave from 8 sine harmonics. More harmonics sharpen the reconstruction and pull it closer to the ideal shape. But at each discontinuity the Gibbs phenomenon leaves a persistent ~9% overshoot spike — adding terms only narrows that ripple, it never removes it, no matter how many harmonics you sum.

Runs locally in your browser — free forever. Scale to the cloud when reality gets heavy.

↓

Kalman Filter

open full solver →

Kalman FilterLive

optimal state estimation

Controls

Presets

Measurement noise

Process noise

The Kalman filter fuses a motion model with noisy measurements to track a hidden state optimally. Each step it predicts, then corrects using the Kalman gain — trusting the measurement more when the model is uncertain, and vice versa. It smooths the jittery sensor (gray) into a clean estimate (cyan) that hugs the truth. The math behind GPS, radar, and spacecraft navigation.

▶ Run in Python

Data Inspector

Measurement RMSE0.0

Kalman RMSE0.0

Noise reductionNaN%

Governing equation

\overset{x}{^} \leftarrow \overset{x}{^} + K (z - H \overset{x}{^}), K = \frac{P H ^{T}}{H P H ^{T} + R}, R = 900, Q = 1.00

Reading this result: Measurement noise dwarfs process noise, so the Kalman gain stays small — the filter leans on its motion model and smooths the jitter hard.

Runs locally in your browser — free forever. Scale to the cloud when reality gets heavy.

↓

Monte Carlo

open full solver →

Monte Carlo Price SimulationLive

geometric Brownian motion

Controls

Presets

Initial price S₀

Expected return μ

Volatility σ

Horizon (years)

Geometric Brownian motion models a price with constant drift and random volatility — the assumption behind Black-Scholes. Running hundreds of simulated paths reveals the full distribution of outcomes, not just an average. Educational tool, not investment advice.

▶ Run in Python

Data Inspector

Median outcome$0

5th percentile$0

95th percentile$0

Governing equation

S_{t + Δ t} = S_{t} e^{(0.07 - \frac{0. 2 ^{2}}{2}) Δ t + 0.2 Δ t Z}, \tilde{S}_{1} \approx $0

Reading this result: Median outcome is about $0, just under the drift-only $107 — GBM compounds in log-space, so volatility always pulls the typical path below the average.

Runs locally in your browser — free forever. Scale to the cloud when reality gets heavy.

or unlock everything with Pro →

About this simulation

The full Time-Series Forecaster tool models a fraud score with the same numerics engineers and scientists use — running entirely client-side. Change any parameter and the result updates in real time, so you can build intuition, check a design, or teach the concept without spreadsheets or installs.

More you can do with Time-Series Forecaster

Time-Series Forecaster for an A/B test Time-Series Forecaster for a survey sample Time-Series Forecaster for a poll margin Time-Series Forecaster for a quality-control chart Time-Series Forecaster for a clinical trial Time-Series Forecaster for a customer churn Time-Series Forecaster for a click-through rate Time-Series Forecaster for an exam-score curve

Other ways to simulate a fraud score

Optimize + UQ for a fraud score Hypothesis Test for a fraud score Linear Regression for a fraud score Central Limit Theorem for a fraud score Confidence Intervals for a fraud score Bayesian Inference for a fraud score

Frequently asked questions

How do I simulate a fraud score?: Open this page and use the live Time-Series Forecaster tool below — set your inputs and the simulation runs instantly in your browser using real numerics. No install, no account needed.
Is it free?: Yes. The simulation runs free in your browser. A one-time unlock or a Pro plan adds advanced parameters, saved presets, data import, and clean exports.
Can I use my own numbers?: Absolutely — every input is adjustable, and with data import you can drive a fraud score from your own measurements.