Mattheus Public Preview

Join Waitlist

Mattheus Public Preview

FEATURED

Why Trading Agents Need Context

FEATURED

Why Trading Agents Need Context

Mattheus Research Notes

Research on AI investing, persistent investor context, and the key reasoning behind each trade.

Search Research

RESEARCH

What Data Lineage Means for Trading Research

Data lineage connects a research result to the source records, transformations, parameters, and people that produced it.

RESEARCH

What Data Lineage Means for Trading Research

Data lineage connects a research result to the source records, transformations, parameters, and people that produced it.

RESEARCH

Why Trading Agents Need Context Across Runs

A useful trading agent needs an inspectable record of goals, evidence, assumptions, tests, corrections, and unresolved questions. A longer chat history cannot supply those relationships by itself.

RESEARCH

Why Trading Agents Need Context Across Runs

A useful trading agent needs an inspectable record of goals, evidence, assumptions, tests, corrections, and unresolved questions. A longer chat history cannot supply those relationships by itself.

MARKET STRUCTURE

From Intelligence to Liquidity

Trading intelligence matters only when an agent can reach fragmented markets without rebuilding execution venue by venue.

MARKET STRUCTURE

From Intelligence to Liquidity

Trading intelligence matters only when an agent can reach fragmented markets without rebuilding execution venue by venue.

AGENT WORKFLOWS

The Continuous Trading Loop

Research, test, monitor, approve, execute, and learn with the same context carrying into the next iteration.

AGENT WORKFLOWS

The Continuous Trading Loop

Research, test, monitor, approve, execute, and learn with the same context carrying into the next iteration.

TESTING

Why Mattheus Starts Paper First

Before live capital enters the loop, paper trading helps validate strategies, permissions, and failure modes.

TESTING

Why Mattheus Starts Paper First

Before live capital enters the loop, paper trading helps validate strategies, permissions, and failure modes.

SECURITY

Guardrails Before Execution

Permissions, risk limits, and approvals keep agentic trading useful without silently expanding authority.

SECURITY

Guardrails Before Execution

Permissions, risk limits, and approvals keep agentic trading useful without silently expanding authority.

RESEARCH

How to Prevent Look-Ahead Bias in Trading Backtests

A backtest is time-valid only when every input and model decision was available at the simulated decision time.

RESEARCH

How to Prevent Look-Ahead Bias in Trading Backtests

A backtest is time-valid only when every input and model decision was available at the simulated decision time.

RESEARCH

How to Model Trading Costs in a Backtest

A useful backtest models the price and execution costs the strategy could have faced when each order was sent.

RESEARCH

How to Model Trading Costs in a Backtest

A useful backtest models the price and execution costs the strategy could have faced when each order was sent.

RESEARCH

How to Detect Backtest Overfitting

A backtest is overfit when its chosen rules explain historical noise better than they support a repeatable trading decision.

RESEARCH

How to Detect Backtest Overfitting

A backtest is overfit when its chosen rules explain historical noise better than they support a repeatable trading decision.

RESEARCH

How to Build a Walk-Forward Test for a Trading Strategy

A walk-forward test trains on past data, evaluates the next untouched period, and repeats that sequence across several decision dates.

RESEARCH

How to Build a Walk-Forward Test for a Trading Strategy

A walk-forward test trains on past data, evaluates the next untouched period, and repeats that sequence across several decision dates.

RESEARCH

How to Estimate Confidence Intervals for a Trading Strategy Backtest

A backtest confidence interval estimates how much a strategy statistic could vary under repeated samples, provided the resampling method preserves the dependence in the data.

RESEARCH

How to Estimate Confidence Intervals for a Trading Strategy Backtest

A backtest confidence interval estimates how much a strategy statistic could vary under repeated samples, provided the resampling method preserves the dependence in the data.

RESEARCH

How to Prevent Survivorship Bias in a Trading Backtest

A survivorship-aware backtest rebuilds the investable universe at every decision date and keeps the returns of securities that later disappeared.

RESEARCH

How to Prevent Survivorship Bias in a Trading Backtest

A survivorship-aware backtest rebuilds the investable universe at every decision date and keeps the returns of securities that later disappeared.

RESEARCH

How to Model Short-Selling Costs in a Trading Backtest

A short-selling backtest needs dated borrow availability, changing fee rates, dividend payments, margin rules, and recall handling before it can estimate an executable return.

RESEARCH

How to Model Short-Selling Costs in a Trading Backtest

A short-selling backtest needs dated borrow availability, changing fee rates, dividend payments, margin rules, and recall handling before it can estimate an executable return.

A cobalt security path separates into price, share, and cash records inside a violet transformation chamber before recombining

RESEARCH

How to Model Corporate Actions in a Trading Backtest

A corporate-action-aware backtest applies each split, dividend, symbol change, merger, and spin-off from a dated event record instead of treating adjusted prices as a complete portfolio ledger.

RESEARCH

How to Model Corporate Actions in a Trading Backtest

A corporate-action-aware backtest applies each split, dividend, symbol change, merger, and spin-off from a dated event record instead of treating adjusted prices as a complete portfolio ledger.

RESEARCH

How to Handle Market Calendars and Time Zones in a Trading Backtest

A reliable backtest converts venue-local sessions into UTC with a dated exchange calendar and preserves the calendar and time-zone versions used for every run.

RESEARCH

How to Handle Market Calendars and Time Zones in a Trading Backtest

A reliable backtest converts venue-local sessions into UTC with a dated exchange calendar and preserves the calendar and time-zone versions used for every run.

RESEARCH

How to Use Point-in-Time Fundamentals in a Trading Backtest

A point-in-time backtest admits each financial fact only after its filing became public, while preserving later amendments as separate versions.

RESEARCH

How to Use Point-in-Time Fundamentals in a Trading Backtest

A point-in-time backtest admits each financial fact only after its filing became public, while preserving later amendments as separate versions.

RESEARCH

How to Model Limit Orders in a Trading Backtest

A limit-order backtest needs venue-specific priority, queue position, partial fills, cancellation timing, and a clear rule for data the simulation cannot observe.

RESEARCH

How to Model Limit Orders in a Trading Backtest

A limit-order backtest needs venue-specific priority, queue position, partial fills, cancellation timing, and a clear rule for data the simulation cannot observe.

RESEARCH

How to Model Trading Halts in a Backtest

A halt-aware backtest separates limit states, regulatory halts, quote resumption, and trade resumption instead of treating the gap as continuous trading.

RESEARCH

How to Model Trading Halts in a Backtest

A halt-aware backtest separates limit states, regulatory halts, quote resumption, and trade resumption instead of treating the gap as continuous trading.

RESEARCH

How to Model Stop Orders in a Trading Backtest

A stop-order backtest separates the trigger event from the market or limit order created after the trigger, then models latency, liquidity, partial fills, and gaps.

RESEARCH

How to Model Stop Orders in a Trading Backtest

A stop-order backtest separates the trigger event from the market or limit order created after the trigger, then models latency, liquidity, partial fills, and gaps.

RESEARCH

How to Model Futures Contract Rolls in a Backtest

A futures backtest should roll between named contracts with a rule fixed in advance, executable prices on both legs, point-in-time liquidity, and product-specific expiration constraints.

RESEARCH

How to Model Futures Contract Rolls in a Backtest

A futures backtest should roll between named contracts with a rule fixed in advance, executable prices on both legs, point-in-time liquidity, and product-specific expiration constraints.

RESEARCH

How to Model Opening and Closing Auctions in a Backtest

An auction-aware backtest separates order acceptance, imbalance messages, freeze periods, the single auction match, and any unfilled quantity from continuous trading.

RESEARCH

How to Model Opening and Closing Auctions in a Backtest

An auction-aware backtest separates order acceptance, imbalance messages, freeze periods, the single auction match, and any unfilled quantity from continuous trading.

Blue and violet glass order blocks form a queue around one warm illuminated order on a dark navy price lane

RESEARCH

How to Model Queue Position in a Trading Backtest

A queue-aware backtest tracks price priority, time priority, executions, cancellations, replacements, and the evidence available at each event.

RESEARCH

How to Model Queue Position in a Trading Backtest

A queue-aware backtest tracks price priority, time priority, executions, cancellations, replacements, and the evidence available at each event.

RESEARCH

How to Model Order Latency in a Trading Backtest

A latency-aware backtest separates decision time, send time, venue receipt, execution, acknowledgment, and market-data arrival before assigning a fill.

RESEARCH

How to Model Order Latency in a Trading Backtest

A latency-aware backtest separates decision time, send time, venue receipt, execution, acknowledgment, and market-data arrival before assigning a fill.