This platform provides academic content only and does not constitute financial or investment
advice of any kind.
Intellectual rigour, applied
How recursive utility theory becomes structured understanding
I came here because I wanted to understand not just what Epstein-Zin preferences imply
for asset prices, but why the mathematics works the way it does and where the empirical
evidence actually stands. The methodology behind this platform answers that question
before the curriculum even begins.
The convictions that shape every explanation published here
Recursive utility theory is not a subject that rewards shortcuts. The separation of risk aversion from intertemporal substitution, the derivation of the stochastic discount factor under Epstein-Zin preferences, the calibration of a long-run risk model — each of these requires following an argument through its full mathematical and economic logic. This platform was built on the conviction that serious learners deserve exactly that: explanations that honour the difficulty of the subject, trace every step back to primary literature, and engage honestly with the empirical debates that remain unresolved. What follows are the philosophical commitments that make that ambition concrete.
1
Every explanation begins with the economic intuition before the mathematics, so that each derivation has a reason rather than a destination.
2
No step in a derivation is described as obvious. If it follows, it is shown to follow, with the intermediate steps made explicit.
3
The primary literature is treated as the authority. Explanations are checked against the original papers of Epstein, Zin, Bansal, Yaron, Hansen, and their collaborators.
4
Where the evidence is mixed — as it genuinely is for long-run risk identification in finite samples — that ambiguity is preserved rather than resolved for the sake of a cleaner narrative.
5
Competing theoretical frameworks are assessed on their own terms before being compared with recursive utility models. Habit formation and rare disaster models are not strawmen.
6
Mathematical approximations — such as the log-linearisation used to solve recursive equilibrium models — are identified as approximations, with their costs and scope made explicit.
7
The platform distinguishes between what a model implies, what the data show, and what researchers currently believe. These are different things and they are kept separate throughout.
From frontier research to structured explanation
Content development follows a disciplined sequence: from identification of the relevant
primary literature through mathematical verification, expert review, and iterative
refinement based on learner engagement.
Primary Literature Identification and Mapping
Every module begins with a systematic mapping of the primary academic literature
relevant to its subject matter. For a module on Epstein-Zin preferences, that
means working from the original 1989 and 1991 papers through to subsequent
extensions, empirical tests, and methodological critiques published in leading
journals. The mapping exercise identifies not only the foundational results but
also the open questions, the contested interpretations, and the methodological
debates that a serious learner will eventually encounter. This stage takes
longer than any other in the development process, because the quality of
everything that follows depends entirely on the accuracy and completeness of the
underlying literature review. No secondary source is treated as authoritative
without verification against the primary text.
Foundational papers mapped
Original Epstein, Zin, Bansal, Yaron, and Hansen papers identified and reviewed against current citation literature.
Open debates catalogued
Unresolved empirical and theoretical questions documented
before any explanatory content is drafted.
Mathematical Verification and Derivation
With the literature mapped, the development team works through every key derivation independently — not transcribing from a source, but reconstructing the argument from first principles to verify that each step follows from the one before it. For the stochastic discount factor module, this means deriving the Epstein-Zin first-order conditions, confirming the relationship between recursive utility and the discount factor, and checking the log-linearisation approximation against the exact solution at relevant parameter values. Where a published paper omits intermediate steps — as many do — those steps are worked out explicitly and their validity confirmed. This process frequently surfaces small errors or ambiguities in the secondary literature that would otherwise propagate into the platform's explanations. Mathematical rigour at this stage is what makes the curriculum trustworthy at the learner's end.
Independent derivation standard
Every key result reconstructed from first principles, not
transcribed, ensuring each logical step is verified.
Approximation costs documented
Log-linearisation and other analytical approximations identified with scope and limitations stated explicitly.
Expert Review by Cuenyzuaslaim Specialists
Completed drafts are reviewed by Cuenyzuaslaim specialists whose expertise spans the relevant theoretical and empirical territory. A module on long-run risk is reviewed by researchers with direct familiarity with the Bansal-Yaron framework and its empirical literature; a module on stochastic discount factors is assessed by economists who work on Hansen-Jagannathan bounds and cross-sectional pricing. Reviewers are asked specifically to identify any step where the explanation departs from the primary literature, any place where the treatment of empirical evidence is unbalanced, and any derivation where an intermediate step is insufficiently justified. Comments are resolved before publication, with substantive disagreements documented and, where necessary, reflected in the explanation itself. The review process is not a formality — it has resulted in substantial revisions to every module published on the platform.
Cuenyzuaslaim specialist review
Each module reviewed by researchers with direct expertise in the relevant theoretical and empirical literature.
Substantive revision standard
Review comments resolved before publication, with
significant disagreements reflected transparently in the
final content.
Learner Feedback and Iterative Refinement
After publication, each module is monitored for learner engagement patterns — specifically, the points at which readers pause, re-read, or seek supplementary material. This information is not used to simplify the content, but to identify where the sequencing of ideas may be creating unnecessary friction. A step that is logically sound but pedagogically abrupt — where the connection between two results is not made sufficiently explicit — will be revised to include additional bridging exposition without altering the underlying mathematics. Feedback from learners who engage substantively with the material is treated as a primary signal: when a practitioner or researcher identifies an explanation that does not align with their reading of the primary literature, that feedback is investigated and, where warranted, incorporated into a revised version. All substantive updates are dated so that returning learners can identify what has changed.
Dated revision tracking
All substantive content updates carry revision dates so learners can identify what has changed since their last visit.
Sequencing refinement
Engagement patterns used to identify pedagogically abrupt transitions, not to simplify the underlying mathematical content.
Three standards, held without exception
These commitments apply to every explanation, every derivation, and every treatment of
empirical evidence published on this platform.
Mathematical Completeness
A derivation that omits steps is not a derivation — it is an assertion. This platform holds to a standard of mathematical completeness that requires every key result to be shown, not merely stated. When the log-linearisation of the recursive equilibrium model is introduced, the approximation is identified, its scope is bounded, and the cost in terms of exactness is made explicit. When the stochastic discount factor is derived under Epstein-Zin preferences, the intermediate steps between the preference aggregator and the first-order conditions are all present. This standard makes explanations longer. It also makes them trustworthy. A learner who follows a complete derivation understands not just that a result holds but why it holds, and that understanding survives an encounter with a paper that modifies the framework in ways a summary cannot anticipate.
Empirical Honesty
Asset pricing theory is not a settled field, and the platform does not present it as one. The evidence for long-run risk — the persistent component in expected consumption growth that drives large risk premiums under recursive preferences — is substantial but contested. Identifying that component in historical data is genuinely difficult, and the debate between long-run risk, habit formation, and rare disaster interpretations of the equity premium remains open. This platform presents those debates as they actually stand in the literature, with the evidence on each side assessed fairly and the unresolved questions acknowledged. Where a model fits some moments of the data well and others poorly, both facts are reported. That kind of honesty is not a limitation — it is what serious engagement with empirical research requires.
Sourcing Transparency
Every theoretical claim and empirical observation on this platform traces back to a named primary source. The foundational results of Epstein and Zin, the long-run risk framework of Bansal and Yaron, the volatility bounds developed by Hansen and Jagannathan — each is attributed precisely, with the original paper identified and the specific result located within it. Where an explanation draws on a secondary synthesis or a textbook treatment, that source is distinguished from the primary literature it represents. This practice serves two purposes: it allows learners to verify any claim against the original text, and it makes clear where the platform is explaining an established result and where it is offering an interpretive account of a contested one. Transparency about sourcing is the foundation on which every other quality standard rests.
Intellectual lineage
From foundational utility theory to modern recursive asset pricing
How to study recursive utility asset pricing effectively
1
Build the dynamic programming foundation before anything else
The Bellman equation and value function iteration underpin every result in recursive utility theory. Learners who arrive without a solid grounding in dynamic programming spend unnecessary effort reconstructing that foundation while simultaneously trying to absorb recursive preferences. Working through the foundations module thoroughly, even if the material feels familiar, pays dividends in every subsequent section where the tools are applied without re-introduction.
2
Read the Epstein-Zin 1989 paper alongside the curriculum
The original Epstein-Zin paper is more readable than its reputation suggests, and
working through it in parallel with the curriculum module builds a habit of verification
that carries through the rest of the programme. When the platform's explanation and
the primary source are read together, each clarifies the other — and the learner
develops the capacity to return to primary literature independently when the curriculum
ends.
3
Treat the log-linearisation as an object of study, not a convenience
The log-linear approximation used to solve recursive equilibrium models is ubiquitous in the long-run risk literature, but its assumptions and costs are rarely examined in detail. Understanding precisely where the approximation binds, what it implies for the accuracy of the equilibrium solution, and when it breaks down gives a learner critical purchase on the entire class of models that rely on it.
4
Engage with the empirical debates before forming a view on the models
The long-run risk model is both influential and contested. Learners who develop a strong
prior about its validity before working through the empirical evidence — the
identification challenges, the finite-sample limitations, the competing explanations —
tend to read that evidence selectively. Approaching the empirical module with genuine
openness about what the data show, and what they cannot show, produces a more durable
and defensible understanding of the framework.
5
Connect every theoretical result to its stochastic discount factor implication
The stochastic discount factor is the common language of asset pricing, and every result
in recursive utility theory ultimately expresses itself through the discount
factor's properties. Building the habit of asking how a change in preferences,
calibration, or model structure alters the discount factor — its level, its volatility,
its co-movement with payoffs — creates a unified analytical framework that applies
across models and asset classes.
