No-Anticipation: Modeling Future-Info Restrictions

Updated 5 July 2026

No-Anticipation Assumption is a modeling restriction that prohibits current states, decisions, or prices from being influenced by future information.
It is applied across diverse fields like DiD, asynchronous systems, portfolio optimization, and causal inference, each adapting the benchmark to its specific intervention and information structure.
Relaxing or refining the assumption exposes hidden causal simplifications, redefines tasks in prediction and control, and highlights the importance of explicitly stating the information structure.

The no-anticipation assumption is a family of domain-specific restrictions that exclude current states, decisions, prices, or outcomes from depending on future information in a prohibited way. Its meaning is not uniform across fields. In some literatures it is a standard adaptedness or causality requirement; in others it is a benchmark assumption whose precise content depends on how interventions, filtrations, rollout schedules, or future event times are specified. Recent work shows that the phrase is often ambiguous unless the relevant information structure is stated explicitly, and that relaxing it can either expose hidden benchmark simplifications or define entirely new tasks and models (Piccininni et al., 17 Jul 2025, 0804.2035, Alonso, 2 Jun 2026).

1. Conceptual scope and formal ambiguity

A central theme in recent literature is that “no anticipation” is not a single theorem-like condition but a modeling choice whose content varies with the object being intervened on. In difference-in-differences, the standard verbal formulation—future treatment has no effect before implementation—can appear redundant in temporally ordered causal models. The ambiguity, according to "Refining the Notion of No Anticipation in Difference-in-Differences Studies" (Piccininni et al., 17 Jul 2025), comes from conflating an intervention on policy implementation with an intervention on the earlier decision/plan/announcement to implement the policy. In asynchronous systems, "The non-anticipation of the asynchronous systems" (0804.2035) reaches a parallel conclusion from a different direction: it introduces several inequivalent causal notions rather than a single universal definition. In decision-theoretic portfolio theory, "Anticipatory Portfolio Optimization" (Alonso, 2 Jun 2026) treats no anticipation as the restricted benchmark in which the optimizer uses a base information set $I_0$ and a fixed reference law, whereas the anticipatory controller solves under richer information $I$ and the deployed law $P_\theta$ .

Domain	Baseline no-anticipation restriction	Representative refinement
DiD	Future implementation should not affect pre-treatment outcomes	Distinguish implementation $A$ from decision/plan $P$ (Piccininni et al., 17 Jul 2025)
Asynchronous systems	Output must not depend on future input	Several nonequivalent definitions, including first-switch and history-based causality (0804.2035)
Portfolio choice	Restricted estimator uses $I_0$ and fixed $\bar P$	Anticipatory optimizer uses richer $I$ and $P_\theta$ (Alonso, 2 Jun 2026)
Finance	No sure profits from future-timed information	Equivalent to absence of predictable jumps (Fontana et al., 2017)

These formulations are technically different, but they share a common role: they specify which future-dependent structures are excluded from admissible reasoning. This suggests that no anticipation is best understood as an information-structure constraint rather than as a single substantive claim.

2. Sequential prediction, perception, and machine anticipation

In action anticipation, recent work shows that standard benchmarks often embed hidden future information. "Untrimmed Action Anticipation" (Rodin et al., 2022) identifies a benchmark assumption that test-time clips are sampled at a fixed offset before the start of a labeled action. In the trimmed formulation, the model predicts the class $y$ of an action starting at time $I$ 0 from the segment beginning at $I$ 1 and ending at $I$ 2. This effectively tells the model when the next action will begin up to the fixed anticipation time $I$ 3. The proposed untrimmed task removes that assumption: at timestamps $I$ 4, sampled every $I$ 5 seconds, the model predicts the set $I$ 6 of actions beginning within a horizon $I$ 7. The EPIC-KITCHENS-100 conversion uses $I$ 8 seconds and $I$ 9 seconds, yielding 1,057,238 training timestamps and 185,532 validation timestamps; 38% of timestamps have no future action in the horizon and 36% contain at least two future action labels. Adapted trimmed baselines remain weak: on action anticipation, RU obtains $P_\theta$ 0 mAP $P_\theta$ 1 and $P_\theta$ 2 mAP $P_\theta$ 3, while RU-5-clf reaches $P_\theta$ 4 mAP $P_\theta$ 5 and $P_\theta$ 6 mAP $P_\theta$ 7, with false positives on 98% of no-action timestamps for the best model. The paper therefore reframes anticipation as a form of future temporal detection requiring answers to whether, what, and when.

The surgical anticipation literature removes a closely related assumption. "Rethinking Anticipation Tasks: Uncertainty-aware Anticipation of Sparse Surgical Instrument Usage for Context-aware Assistance" (Rivoir et al., 2020) criticizes setups in which a new action is assumed to occur within typically one second and only its category is unknown, or in which dense temporal segmentations are required during training and inference. Instead, it defines, for each frame $P_\theta$ 8 and instrument $P_\theta$ 9, a truncated remaining-time target

$A$ 0

with $A$ 1 minutes, together with an auxiliary three-way label $A$ 2. At inference the method uses only image data, without dense action segmentation, phase labels, instrument presence annotations, future temporal boundaries, or knowledge of final surgery duration. This is a strict no-future-timing formulation at test time, even though target construction uses future occurrences offline during training.

Other recent work questions whether no anticipation should be identified with dense temporal video processing. "Understanding Multimodal Complementarity for Single-Frame Action Anticipation" (Benavent-Lledo et al., 29 Jan 2026) studies prediction of the action occurring $A$ 3 ahead from a single RGB frame, optionally depth, and semantic history. The resulting AAG+ framework improves the original AAG and, on IKEA-ASM, reaches $A$ 4 top-1/top-5 compared with VLMAH at $A$ 5; on Assembly101, however, AAG+ remains below AVT and VLMAH on Recall@5, indicating that dense temporal context still matters in higher-ambiguity settings. The paper therefore challenges the assumption that explicit temporal video input is always necessary, but not the need for structured context.

A different boundary appears in long-term multimodal anticipation. "Multi-level and Multi-modal Action Anticipation" (Kim et al., 3 Jun 2025) operates on partially observed video prefixes, yet its segmentation and multimodal transformers use unmasked self-attention and cross-attention, so it is best characterized as prefix-based at the sequence level rather than strictly causal within the prefix. By contrast, "Zero-Shot Anticipation for Instructional Activities" (Sener et al., 2018) keeps the test-time constraint explicit: given ingredients and an observed text or video prefix, it predicts subsequent steps before seeing those segments, using external instructional corpora to learn procedural structure offline.

3. Causal inference and experimental design

In causal inference, the modern literature increasingly treats no anticipation as a question about the correct intervention, not merely about temporal ordering. In a two-period DiD setup, "Refining the Notion of No Anticipation in Difference-in-Differences Studies" (Piccininni et al., 17 Jul 2025) notes that standard DiD identifies

$A$ 6

under positivity, consistency, and parallel trends formulated with respect to $A$ 7, without any extra no-anticipation restriction. The paper then enlarges the model with a decision variable $A$ 8, where $A$ 9, and redefines no anticipation as

$P$ 0

Under parallel trends with respect to $P$ 1, consistency for $P$ 2, and an exclusion restriction on direct effects of $P$ 3 on $P$ 4, the classic DiD contrast equals $P$ 5, where

$P$ 6

Under Assumptions 5–7, the same contrast identifies

$P$ 7

The paper’s main conceptual claim is that the usual subject-matter concern is typically about $P$ 8, not $P$ 9.

"A Joint Analysis of Sensitivity to Anticipation and Parallel Trends Violations" (Fenaroli, 1 Mar 2026) pushes this further by showing that observed pre-trends confound anticipation and parallel-trends failure. Using

$I_0$ 0

for untreated-trend deviations and

$I_0$ 1

for anticipation effects, the paper derives

$I_0$ 2

and

$I_0$ 3

Thus a pre-treatment event-study coefficient $I_0$ 4 is not purely evidence about parallel trends unless $I_0$ 5. The paper parameterizes anticipation increments as $I_0$ 6 with $I_0$ 7, or alternatively as $I_0$ 8 with $I_0$ 9, and derives sharp identified sets for $\bar P$ 0. It also shows that the identified set under joint deviations can be shorter than the set under parallel-trends violations alone only when anticipation is constrained away from zero; if $\bar P$ 1, that shortening cannot occur.

In stepped wedge cluster randomized trials, no anticipation is embedded directly in the analysis model. "On Anticipation Effect in Stepped Wedge Cluster Randomized Trials" (Wang et al., 10 Apr 2025) shows that the standard Hussey–Hughes model

$\bar P$ 2

and the exposure-time interaction model

$\bar P$ 3

both encode the restriction that treatment-related effects begin only when $\bar P$ 4. The anticipation-augmented forms add $\bar P$ 5: $\bar P$ 6 and

$\bar P$ 7

If the true model is HH-ANT but HH is fit, then

$\bar P$ 8

so omission of anticipation biases the treatment estimate. The paper also proves $\bar P$ 9 and $I$ 0, implying that modeling anticipation increases variance and that ignoring it in sample-size planning can lead to underpowered trials.

4. Finance, arbitrage, and anticipatory value

In mathematical finance, no anticipation is often formalized through admissible information and the exclusion of exploitable predictable price moves. "On the existence of sure profits via flash strategies" (Fontana et al., 2017) studies a càdlàg adapted gains process $I$ 1 with jump process $I$ 2. Its central result is that there are no sure profits via flash strategies if and only if $I$ 3 does not exhibit predictable jumps, and no constant profits via flash strategies if and only if $I$ 4 does not exhibit fully predictable jumps. Economically, this means that price moves at predictable dates may occur, but their direction cannot be known in advance if one wants to exclude sure high-frequency arbitrage; if the size is also known, even stronger arbitrage results. The paper therefore concludes that price changes at scheduled dates should only be due to unanticipated information releases.

"Convergence of the financial value of weak information for a sequence of discrete-time markets" (Lindsell, 2022) relaxes the no-anticipation benchmark more conservatively. Trading strategies remain predictable: $I$ 5 and in continuous time the admissible class $I$ 6 consists of $I$ 7-predictable processes. The relaxation occurs at the level of beliefs: instead of revealing the exact future signal $I$ 8, the agent knows only its law $I$ 9. The associated minimal probability measure is

$P_\theta$ 0

with $P_\theta$ 1, $P_\theta$ 2. The financial value of weak information is then

$P_\theta$ 3

This is a relaxation of no anticipation in the agent’s information set, but not in the admissibility of controls.

The finance literature also distinguishes between the possibility of anticipation and its equilibrium expression. "Does the Market Anticipate? Can it? Should it?" (Wren, 2 Mar 2026) studies a continuous-time setting with pre-horizon risk resolution. The model allows public predictive data $P_\theta$ 4 and posterior beliefs $P_\theta$ 5, so pre-disclosure anticipation is formally possible. Yet the paper argues that optimized trading can suppress anticipation of predictable risk outcomes. In particular, with relative information intensity

$P_\theta$ 6

low-RII means $P_\theta$ 7, and under that condition the heterogeneous-market clearing price can have a steady-state Status Quo Bias, with $P_\theta$ 8. The implication is that underreaction to predictable risks need not contradict no-arbitrage.

"Anticipatory Portfolio Optimization" (Alonso, 2 Jun 2026) generalizes the benchmark across insider information, dynamic planning, and market impact. The restricted policy is

$P_\theta$ 9

whereas the anticipatory policy is

$y$ 0

For permanent linear impact, the price-taking allocation is

$y$ 1

while the impact-aware allocation is

$y$ 2

The value of anticipation is the realized control gap $y$ 3. The paper further shows that correctly specified anticipation creates value, vacuous anticipation has zero value, and misspecified anticipation is harmful, with exact penalty

$y$ 4

5. Control, optimization, asynchronous systems, and collective dynamics

Outside finance, no anticipation often appears as a control-theoretic or systems-theoretic causality restriction. "Online Proactive Multi-Task Assignment with Resource Availability Anticipation" (Nedelmann et al., 2023) is explicit that it adopts no anticipation only on the demand side. The method assumes no information on upcoming requests, no arrival distribution, and no assumption concerning future locations of requests; however, it does anticipate which currently busy agents will become available within a receding horizon $y$ 5. The resulting setting is therefore partial anticipation: demand-side no anticipation, supply-side anticipation.

In stochastic analysis, "Infinite Anticipation Backward Stochastic Differential Equations" (Cheng et al., 19 Nov 2025) removes the standard no-future-dependence structure from the BSDE generator while retaining adapted solutions. The new IABSDE takes the form

$y$ 6

Under a path-Lipschitz condition (H1) and integrability condition (H2), the paper proves existence and uniqueness of an $y$ 7-adapted solution in $y$ 8, and derives a comparison result. The paper interprets this as a dual counterpart to SDEs with infinite delay, showing that standard no-anticipation in the driver is not mathematically necessary if adaptedness of the generator value and suitable weighted estimates are preserved.

In classical systems theory, "The non-anticipation of the asynchronous systems" (0804.2035) develops several inequivalent formalizations for asynchronous circuits. The simplest one says that if $y$ 9 and $I$ 00 are both variable, then

$I$ 01

so the first output switch cannot precede the first input switch. A stronger, history-based version requires

$I$ 02

The paper then adds bounded-memory and delayed-window variants, and proves structural results such as preservation under subsystem restriction and time-shift normalization.

A distinct use of anticipation appears in collective-motion models. "Anticipation induces polarized collective motion in attraction based models" (Strömbom et al., 2017) contrasts the baseline assumption—each particle updates from current neighbor positions or headings—with a model in which the local center of mass is computed from anticipated positions. In the synchronous Local Attraction Model without anticipation, the regime diagram is: no group for $I$ 03, mills for $I$ 04, and swarms for $I$ 05. Adding anticipation causes cohesive polarized groups to emerge for $I$ 06 and again for sufficiently large $I$ 07, despite the model having no explicit alignment term. In this setting, relaxing the no-anticipation assumption qualitatively changes the phase behavior rather than merely refining estimation.

6. Cross-domain implications and open directions

Across these literatures, relaxing no anticipation rarely amounts to a uniform weakening of causality. More often, it exposes a hidden benchmark assumption that had already encoded privileged alignment, timing, or information. In action anticipation, knowing that evaluation clips are taken exactly $I$ 08 seconds before onset turns prediction into a materially easier task than continuous-video forecasting (Rodin et al., 2022). In DiD, an apparently standard no-anticipation formula may be either redundant or misdirected unless the intervention on $I$ 09 is separated from the earlier decision variable $I$ 10 (Piccininni et al., 17 Jul 2025). In finance, predictable dates are admissible, but predictable jump direction or magnitude is not if flash-strategy arbitrage is to be excluded (Fontana et al., 2017).

The literature also shows that “more anticipation” is not uniformly better. Single-frame action anticipation can be competitive when procedural state and semantic history are highly constraining, but its advantage weakens in more ambiguous, high-variability settings (Benavent-Lledo et al., 29 Jan 2026). Anticipatory portfolio optimization yields positive value under correct specification, zero value under vacuous enrichment, and a precise overfitting penalty under misspecification (Alonso, 2 Jun 2026). Market microstructure models even suggest that optimized trading may suppress anticipation rather than express it immediately (Wren, 2 Mar 2026). This suggests that the substantive question is not whether a model is anticipatory in the abstract, but which future-dependent structures are admissible, identifiable, and decision-relevant in a given domain.

Open problems differ by field but are structurally similar. Untrimmed video prediction highlights the need to model no-action periods, multiple future events, and accurate time-to-action estimation (Rodin et al., 2022). DiD sensitivity analysis shows that pre-trends need joint treatment as mixtures of anticipation and untreated-trend deviations rather than as one-dimensional diagnostics (Fenaroli, 1 Mar 2026). Stepped wedge trials require models that accommodate anticipation together with exposure-time heterogeneity and updated power calculations (Wang et al., 10 Apr 2025). Stochastic analysis now admits infinite anticipation in BSDE generators, but only under strong measurability, Lipschitz, and integrability conditions (Cheng et al., 19 Nov 2025). Taken together, these works indicate that the no-anticipation assumption remains indispensable as a benchmark, but only after its information structure, intervention target, and admissibility role have been specified with technical precision.