Synthetic demonstration data

Auditable analytics for collective financing

An analytical methodology for turning participant behavior into forward-looking financial decisions.

MarkovIQ is a specialized decision engine, not a CRM. Its transparent algorithms predict what participants may do next, trace the effect through group cash flows, and show why management attention is needed.

Explore the interactive demo See how every result is checked

Behavior-to-cash-flow traceVersioned model evidenceBase and stress views

Executive portfolio

Reserve pressure overview

Run 5F1C5433

Covered groups

7 / 22

Fund balance

LCU 10M

High-risk queue

567

G-006 · Priority 01

watch · 46

Early-tenure cancellation behavior translates into a 100% 24-month reserve-breach probability.

Healthy comparison

G-021 · Health 86

0% reserve risk

Behavioral effect

46.9 expected exits

Lost contributions + refund pressure

Built for risk, finance, and liquidity leaders in collective financing.

Question → Data → Model → Evidence → Decision

One connected analytical chain

Don’t just predict it. Demonstrate why.

MarkovIQ connects behavioral uncertainty to the economics of a financing group, with a visible evidence trail at every step.

Predict participant behavior

Estimate cancellation risk from tenure, payment behavior, recent cure, and financial indicators.

Propagate expected states

Translate participant-level probabilities into expected group behavior over time.

Project group cash flows

Connect contributions, refunds, allocations, and obligations to a reconciled fund forecast.

Prioritize a decision

Show where reserve pressure may emerge, why, and which queue deserves a policy-compliant review.

Methodology, not recordkeeping

Start with the participant, then follow the money.

A cancellation probability is useful only when leaders can see what it means for expected contributions, refund pressure, future obligations, and minimum reserve coverage. MarkovIQ applies a governed analytical chain instead of adding another customer-management workflow.

Prioritized participant queue with directional drivers
Expected cash-flow bridge that always reconciles
Fixed-seed forecast range and reserve-breach definition
Explicit model coverage and incomplete-data handling

Follow the featured G-006 story

Participant evidence

See the current value, reference cohort, direction, and probability-point effect of every driver.

Cash-flow propagation

Connect expected exits to lost contributions and refund pressure before viewing group reserves.

Transparent health

Inspect each reading, floor, target, sub-score, weight, and contribution to the index.

Governed calculation

Keep AI out of the financial math. Every output comes from a versioned deterministic engine.

Base versus combined stress

Make assumptions visible before discussing outcomes.

Compare the expected path with one disclosed stress: cancellation +20%, delinquency +15%, obligations +10%, and advance-payment participation −15%.

Compare the scenarios

Base case

Expected path

Observed behavior and contractual assumptions.

Combined stress

Same engine

Changed assumptions, persisted with the run.

Built by QED Analytica

Financial engineering meets behavioral AI.

The company brings together quantitative finance, risk, machine learning, and production technology—with banking experience across the United States, Mexico, and Colombia—to build specialized forward-looking decision systems.

Alex Restrepo

Co-Founder · Quantitative Finance & Risk

Three decades across banking, capital markets, mortgage finance, cash-flow modeling, valuation, liquidity, stress testing, and analytical systems.

Alejandro Correa Bahnsen, PhD

Co-Founder · AI, Credit Risk & Technology

Data and AI executive, entrepreneur, professor, and researcher focused on behavioral modeling, credit risk, explainability, and production AI products.

Follow one risk signal all the way to a financial decision.

Explore a fully synthetic, reproducible workflow with no client data and no black-box claims.

Open MarkovIQ