who we are

We started this because every dividend model we encountered assumed certainty where none exists, and that struck us as intellectually dishonest

team of quantitative researchers reviewing dividend models on screens
growth path

Slow compounding of knowledge over eleven years of focused research

35%

average model refinement rate

We are a small group of quantitative researchers and former practitioners who build stochastic frameworks for modeling uncertain future dividend streams. Our work sits at the intersection of probability theory, corporate finance, and equity valuation. We do not sell courses. We do not manage money. We document what we have learned, publish our methodologies, and invite scrutiny. The field of dividend growth modeling has long relied on deterministic projections that ignore path-dependency and regime shifts. We think that is a problem worth solving.
see our methods
abstract mathematical formulas on dark background

We believe that a model you cannot break is a model you do not understand, so we spend most of our time trying to break our own

1

mathematical rigor

Every framework we publish begins with a clearly stated set of assumptions and ends with a proof of convergence or an explicit bound on error. We do not hand-wave. If we cannot derive it, we do not present it as settled.

2

practical use

A model that lives only in a journal is a model that dies there. We test every framework against historical dividend paths, document where it fails, and publish the failure modes alongside the successes.

3

intellectual honesty

We revise publicly. When new data contradicts an earlier framework, we update it, date the revision, and explain what changed. The goal is not to be right first. The goal is to be less wrong over time.

what guides our work

These are not aspirational. They are operational constraints we apply to every model and every publication.

1

academic integrity

We cite sources we have read. We distinguish between our own derivations and results we are borrowing. We do not inflate significance or suppress contradictory findings.

2

accessibility

Complex ideas should be explained clearly, not hidden behind jargon. We write for the curious beginner who is willing to work, not for the specialist who already knows.
3

real-world relevance

Every framework must connect to an actual valuation problem. If we cannot map it to a decision a practitioner faces, we do not publish it.

4

field advancement

We contribute back. When we improve on an existing method, we publish the improvement under an open license so others can build on it.

5

continuous revision

No model is final. We maintain versioned documentation, track changes over time, and treat every framework as a living document subject to update.

the people building these frameworks

We are researchers first. Some of us came from practice, others from academia. What unites us is a shared intolerance for models that look elegant but fail under stress. We work remotely across time zones and meet rarely. Most of our collaboration happens in writing.

Dr. Eleanor Vance

Dr. Eleanor Vance

Lead researcher, stochastic methods

Wellington, New Zealand

core competencies

stochastic calculus dividend modeling python programming peer review

Eleanor spent fifteen years in quantitative equity research before turning to full-time methodology development. She leads our stochastic dividend growth frameworks and insists that every model include a stress-test protocol.

Marcus Chen-Whitmore

Marcus Chen-Whitmore

Senior researcher, empirical validation

Auckland, New Zealand

core competencies

historical backtesting statistical inference data architecture model validation

Marcus designs the empirical tests that separate robust frameworks from curve-fitted artifacts. He maintains our historical dividend database and publishes the validation reports that accompany each new model release.

Aroha Te Kanawa

Aroha Te Kanawa

Research fellow, regime detection

Christchurch, New Zealand

core competencies

regime switching time series analysis bayesian methods documentation

Aroha specializes in detecting structural breaks in dividend growth patterns. Her work on hidden Markov approaches to regime detection has been integrated into three of our published frameworks.

James Okonkwo

James Okonkwo

Research fellow, computational implementation

Dunedin, New Zealand

core competencies

numerical methods code optimization reproducible research open source

James translates theoretical models into reproducible computational implementations. He ensures that every published framework ships with working code, documented assumptions, and a clear path to replication.

Our history is a sequence of research decisions, not funding rounds or product launches. Each milestone marks a shift in what we thought was possible.

how we got here

Est. 2015 5 entries
research
2015

initial research focus

We began as an informal reading group studying dividend discount models. The deterministic assumptions in standard textbooks bothered us. We started collecting historical dividend data to test whether the models matched reality.

publication
2017

first stochastic framework

After two years of data collection, we published our first stochastic dividend growth model. It was crude, assumed log-normal growth, and failed during the 2008 stress period. We published the failure openly.

methodology
2019

regime detection added

Aroha joined and introduced hidden Markov regime switching. This allowed our models to adapt when dividend growth patterns shifted structurally. Validation against historical data improved measurably.

platform
2021

open documentation launched

We moved all our frameworks to versioned, open documentation. Every model now includes derivation, code, validation report, and known limitations. We stopped treating publications as endpoints.

community
2024

professional community formed

Practitioners began using our frameworks in their own work and contributing feedback. We formalized our structure to support ongoing collaboration without becoming a consultancy or a training provider.

Cookie notice

We use cookies

This website uses cookies to function properly and to help us understand how visitors interact with our research content.

Cookies are small text files stored on your device. We use essential cookies for website functionality and analytical cookies to understand usage patterns. We do not use cookies for targeted advertising. You can manage your preferences at any time.

Privacy policy