Skip to content
TP Advisory Services

QISTRAInternal quantitative frameworkQuantitative validation for investment strategies and actively managed solutions.

QISTRA is a proprietary analytical framework for index engineering, strategy validation, manager analysis and investment-process assessment.
QISTRA
  • Strategy Backtesting
  • Factor Attribution
  • Skill vs Luck
  • Manager DNA
  • Decision Analysis
  • Portfolio Reconstruction

Benchmark Comparison

Comparison against a benchmark chosen for its relevance to the strategy rather than for the picture it produces.

Index Engineering

Design, specification and testing of custom indices, including rule definition, rebalancing logic and robustness checks.

  1. 01Strategy rules
  2. 02Reconstruction
  3. 03Attribution
  4. 04Significance
  5. 05Assessment

Modules

What the environment does.

Each module below is part of QISTRA and is described exactly as it is used in client work.
  • 01

    Strategy Backtesting

    Rule-based strategies rebuilt from their definition and tested with explicit cost, timing and implementation assumptions.

  • 02

    Factor Attribution

    Return decomposed into known factor exposures, so what is systematic is separated from what is specific to the manager.

  • 03

    Skill vs Luck

    Statistical assessment of whether an observed track record is distinguishable from a sample drawn by chance.

  • 04

    Manager DNA

    The behavioural signature of a process — turnover, concentration, sizing, persistence and how consistently rules are applied.

  • 05

    Decision Analysis

    Individual decisions examined for their contribution, so that outcome and reasoning are not confused with each other.

  • 06

    Portfolio Reconstruction

    Historical positions and weights rebuilt from available data when the underlying holdings are incomplete or undisclosed.

  • 07

    Benchmark Comparison

    Comparison against a benchmark chosen for its relevance to the strategy rather than for the picture it produces.

  • 08

    Index Engineering

    Design, specification and testing of custom indices, including rule definition, rebalancing logic and robustness checks.

Method

Five things a track record confuses

The purpose of the framework is to separate them, and to say how confidently each can be established from the data available.
  1. 01

    Performance

    What was actually realised, net of costs, over the period observed — the starting point, not the conclusion.

  2. 02

    Factor exposure

    The part of the return explained by systematic exposures that can be obtained elsewhere, at a lower fee.

  3. 03

    Management decisions

    The contribution of the choices actually made: sizing, timing, entries, exits and deviations from the rules.

  4. 04

    Luck

    The portion of the outcome that a sample of this length and volatility cannot distinguish from chance.

  5. 05

    Repeatable skill

    What remains once the four above are accounted for — and the honest confidence interval around it.

Separating what happened from what can be repeated

A track record mixes several things at once. Part of it comes from broad market exposure, part from factor tilts that can be replicated cheaply, part from decisions the manager actually made, and part from the particular sample of history that happened to occur.

QISTRA is built to take those apart. The objective is not to produce a score, but to state clearly what portion of performance is explained by exposure, what portion is attributable to the investment process, and what portion cannot be distinguished from chance given the length and volatility of the record.

For allocators, product teams and managers themselves

The same framework serves three different questions. An allocator wants to know whether a process is worth its fee. A product team wants to know whether a strategy is robust enough to be indexed or wrapped. A manager wants an independent reading of their own process.

In each case the deliverable is the analysis and the reasoning behind it, in a form that can be reviewed and challenged.

Interested in what QISTRA can answer for your book?

QISTRA is a proprietary analytical framework for index engineering, strategy validation, manager analysis and investment-process assessment.