LEARN / DECISION PATHS

Learn research decisions, not button sequences.

kquant learning paths follow real tasks: find a mechanism, read an outcome, detect overfitting, launch Simulation and diagnose divergence.

Launch research Explore the details ↓

SYSTEM DOMAIN07 / TRUST

Keep product rules and boundaries visible

DISCOVER
OPTIMIZE
VALIDATE
RUN
OBSERVE

Getting started

Turn a market idea into a minimal testable rule.

Backtesting

Understand look-ahead, next-bar logic, costs and trade count.

Overfit controls

Learn holdout, DSR, PBO, walk-forward and Monte Carlo.

Simulation and live

Connect an exchange securely and know when to stop execution.

DECISION FRAMEWORK

Which boundary protects the user here?

Product, legal and security rules must describe the same capability, responsibility and limitation without promising an unavailable guarantee.

Use precise terms

Expose limitations

Keep an independent stop path

OUTPUT

Trust record

Starter collection

Four routes for your first research cycle

Why backtests lie

Look-ahead, selection bias and idealized fills

DSR in plain English

How trial count changes confidence in Sharpe

Simulation vs live

What simulation can and cannot test

Secure API access

Permissions, subaccounts and IP whitelists

Important

Historical, statistical and model results do not promise future returns. kquant provides computational tools and technical execution, but does not assess whether a specific trade is suitable for a user.