phenomenon · claim-type · cross-domain-transfer · mechanism · claim-relation · replication-status · limitation — grounded in the full text| phenomenon | heterogeneous treatment effects— “In this paper we study two closely related problems: first, estimating heterogeneity by covariates or features in causal effects in experimental or observational studies, and second, conducting inference about the magnitude of the differences in treatment effects across subsets of the population.” | |
| claim-type | method— “In this paper we propose methods for estimating heterogeneity in causal effects in experimental and observational studies, and for conducting hypothesis tests about the magnitude of the differences in treatment effects across subsets of the population.” | |
| method — evidence | — “To assess the relative performance of the proposed algorithms we carried out a small simulation study with three distinct designs.” | |
| transfer | from statistics / machine learning: regression-tree (recursive partitioning) methods adapted to estimate heterogeneous treatment effects— “Our approach builds on regression tree methods, modified to optimize for goodness of fit in treatment effects” | transfers[0] |
| mechanism | honest estimation: use one sample to construct the tree partition and a separate sample to estimate treatment effects, removing overfitting bias— “one sample is used to construct the partition and another to estimate treatment effects for each subpopulation” | mechanisms[0] |
| builds-on | regression tree methods (machine learning)— “Our approach builds on regression tree methods” | claim_relations[0] |
| builds-on | standard CART regression trees ([4], [3])— “In this paper, we focus on the analogous goal of deriving a partition of the population according to treatment effect heterogeneity, building on standard regression trees ([4], [3]).” | claim_relations[1] |
| builds-on | the Rubin Causal Model / potential outcome framework ([19], [11], [14])— “Causal effects, in the Rubin Causal Model or potential outcome framework we use here ([19], [11], [14]), are comparisons between outcomes we observe and counterfactual outcomes we would have observed under a different regime or treatment.” | claim_relations[2] |
| sense | 'honest' - A STIPULATED PROPERTY OF DATA USE: a model is honest when the information selecting its structure is not the information estimating its parameters, which is sample splitting and nothing about truthfulness, candour or good faith - all of which the word carries anyway— “We say that a model is “honest” if it does not use the same information for selecting the model structure (in our case, the partition of the covariate space) and for estimation given a model structure” | senses[0] |
| sense | 'adaptive' - THE NAME GIVEN TO THE ALTERNATIVE: conventional CART, where one sample does both jobs, is labelled adaptive here so that the two can be compared; the word is the paper's, applied to a practice that did not call itself that— “We contrast ‘’honest” estimation with “adaptive” estimation used in conventional CART, where the same data is used to build the partition and estimate leaf effects” | senses[1] |
| sense | the estimand - UNOBSERVABLE FOR EVERY UNIT: unit-level causal effects are not observed for any unit at all, so unlike an ordinary prediction problem there is no held-out truth anywhere in the data against which a per-unit estimate could be scored— “This creates complications for conventional methods because we do not observe unit level causal effects for any unit” | senses[2] |
| control | three simulation designs differing in whether noise covariates are present and in how covariates enter the treatment effect — held— “We find in the settings we consider that honest estimation results in improvements in fit, in some cases very large improvements, over a more traditional “adaptive” estimation approach.” — rules out that the advantage of honest estimation depends on one particular data-generating design, in particular on the absence of noise covariates— “Design 1 does not have noise covariates.” | controls[0] |
| retracted | EVIDENCE CORRECTED (RE-GROUNDED) · honest estimation —contrasted-with→ adaptive — RE-GROUNDED. The old span named only one endpoint. This sentence names both and states the contrast WITH ITS DIRECTION - honest estimation improves on adaptive, not merely differs.— “We contrast ‘’honest” estimation with “adaptive” estimation used in conventional CART, where the same data is used to build the partition and estimate leaf effects”— “We contrast ‘’honest” estimation with “adaptive” estimation” | concept_edges[0] |
| note | honest estimation —contrasted-with→ adaptive — RE-GROUNDS another row of this record: the old span named only ONE endpoint.— “We find in the settings we consider that honest estimation results in improvements in fit, in some cases very large improvements, over a more traditional “adaptive” estimation approach.” | concept_edges[2] |
| outcome | control — held— “Design 1 does not have noise covariates.”— “We find in the settings we consider that honest estimation results in improvements in fit, in some cases very large improvements, over a more traditional “adaptive” estimation approach.” | controls[0] |
| for which use | limitation — the precision of honest estimates: THERE IS A LOSS OF PRECISION DUE TO SAMPLE SPLITTING, offset in fit by the elimination of bias— “Although there is a loss of precision due to sample splitting (which reduces sample size in each step of estimation), there is a benefit for fit in terms of eliminating bias that offsets at least part of the cost.” | limitations[1] |
| for which use | limitation — the recommendation itself, with the cost measured: THERE IS A COST TO HONEST ESTIMATION IN TERMS OF MSE-tau, VARYING BY DESIGN AND ESTIMATOR— “The results show that there is a cost to honest estimation in terms of MSEτ , varying by design and estimator.” | limitations[2] |
| for which use | limitation — any application where units interact: the no-interference assumption MAY BE VIOLATED IN SETTINGS WHERE SOME UNITS ARE CONNECTED THROUGH NETWORKS— “This assumption may be violated in settings where some units are connected through networks.” | limitations[3] |
| for which use | limitation — the hybrid variants: WE DO NOT CONSIDER THOSE VARIATIONS IN THIS PAPER— “We do not consider those variations in this paper.” | limitations[4] |
| for which use | limitation — any treatment-effect estimate produced by this method: unit-level causal effects are not observed for any unit, so the target is never available to check an estimate against— “This creates complications for conventional methods because we do not observe unit level causal effects for any unit” | limitations[5] |
| for which use | limitation — the word 'honest' as used throughout this paper: a model is honest if it does not use the same information for selecting the model structure and for estimation given that structure— “We say that a model is “honest” if it does not use the same information for selecting the model structure (in our case, the partition of the covariate space) and for estimation given a model structure” | limitations[6] |
| for which use | limitation — the benefit reported for honest estimation: it is an improvement in confidence-interval coverage, attaining nominal rates, and the authors describe the cost as not much sacrifice in fitting treatment effects— “we show that honest estimation can result in substantial improvements in coverage of confidence intervals, where our method attains nominal coverage rates, without much sacrifice in terms of fitting treatment effects” | limitations[7] |
| for which use | limitation — the contrast drawn here between honest and adaptive estimation: adaptive is the label the authors give conventional CART, in which the same data builds the partition and estimates leaf effects— “We contrast ‘’honest” estimation with “adaptive” estimation used in conventional CART, where the same data is used to build the partition and estimate leaf effects” | limitations[8] |
| replication | untested | |
| limitation | requires the unconfoundedness (conditional randomization) assumption— “we maintain the assumption of randomization conditional on the covariates” | limitations[0] |
| limitation | Sample splitting for honest estimation costs precision, which is only partly offset by the removal of bias.— “Although there is a loss of precision due to sample splitting (which reduces sample size in each step of estimation), there is a benefit for fit in terms of eliminating bias that offsets at least part of the cost.” | limitations[1] |
| limitation | In the simulations honest estimation is measurably worse than adaptive estimation on treatment-effect mean-squared error.— “The results show that there is a cost to honest estimation in terms of MSEτ , varying by design and estimator.” | limitations[2] |
| limitation | The no-interference (SUTVA) assumption the methods rest on can fail when units are connected in a network. | limitations[3] |
| limitation | Hybrid variants (adaptive tree construction combined with honest leaf estimation) are left unexamined. | limitations[4] |
| limitation | The quantity the method estimates is never observed for any unit in the data. | limitations[5] |
| limitation | The paper's central adjective is a stipulated property of data use, not a claim about candour. | limitations[6] |
| limitation | The improvement claimed for honesty is measured in coverage, and the paper names what it costs. | limitations[7] |
| limitation | The comparison class is conventional tree estimation under a name the paper assigns it. Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | limitations[8] |
finding · quantity · variable · definition · key-term — grounded in the full text| finding | A data-driven approach partitions the data into subpopulations that differ in the magnitude of their treatment effects, letting the data discover relevant subgroups. Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.findings[0] |
| not machine-vouched | finding — declared-ellipsis - An elided quote: its 2 pieces ('We provide a data- driven approach to pa' ... 'let the data discover relevant subgroups') are each in our copy of the paper (with line-break hyphens, spacing and punctuation set aside), and the quote marks the cut with an ellipsis, so the joined string is in no copy. Located mechanically, piece by piece. | content_tags.findings[0] |
| finding | Honest estimation splits the sample: one half builds the partition, the other estimates each subpopulation's treatment effect. Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.findings[1] |
| variable | covariates (independent) Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.variables[0] |
| variable | treatment effects (dependent) Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.variables[1] |
| definition | honest estimation Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[0] |
| definition | heterogeneous treatment effects Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[1] |
| evidence status note | definition — QUOTE REPAIRED: the extractor split a hyphen-space and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. | content_tags.definitions[1] |
| key-terms | treatment, causal, effects, partition, honest, tree, covariates, estimation, subgroup |
| concept | honest estimation Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[4] |
| concept | adaptive Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[5] |
| concept | unconfoundedness Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[6] |
| concept | mean-squared error Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[7] |
| concept | coverage Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[8] |
| Bayesian | Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[0] |
| confidence-interval CI | as written: confidence intervals Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[1] |
| false-discovery-rate FDR | as written: multiple testing Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[2] |
| hypothesis-testing | as written: hypothesis tests Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[3] |
| evidence status note | hypothesis-testing — QUOTE REPAIRED: the extractor split a hyphen-space and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. | content_tags.statistics[3] |
| mean μ | as written: mean of Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[4] |
| null-hypothesis H0 | Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[5] |
| evidence status note | null-hypothesis — THE QUOTE STOPS WHERE THE EXTRACTED TEXT GOES OUT OF ORDER: the extraction splices another line of the page ('If the two leaves are denoted L (Left) and') into this sentence, so the quote ends before that line, at 'the two potential', and the end of the sentence is not quoted. | content_tags.statistics[5] |
| p-value p | as written: p-values Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[6] |
| regression β | Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[7] |
| evidence status note | regression — QUOTE REPAIRED: the extractor split a hyphen-space and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. | content_tags.statistics[7] |
| sample-size n/N | Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[8] |
| t-test t | as written: T-statistic Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[9] |
| variance σ² | Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[10] |
When we have changed our mind. Some entries in the table begin with a word in capitals and the reason for it. EVIDENCE CORRECTED — the finding still stands, but we had quoted the wrong sentence from the paper, and the quotation has been replaced with the right one. RETRACTED — the entry itself was wrong; it stays on the page, marked, with the reason, rather than being deleted. WITHDRAWN AND REPLACED — the entry was wrong, and a corrected version of it appears in the entries below. Each mark carries its own reason, in brackets, because the reasons differ from one entry to the next. An entry we have WITHDRAWN is removed from this page and kept in the record, with its reason. A line reading Audit and a date is a later re-reading of the entry against its whole sentence: DOUBTFUL means it was kept, and the reason says what is loose about it. Quotation marks inside the tagger's sentence. Where a tag is the tagger's own sentence, a phrase in quotation marks inside it is borrowed from the paper — the tagger's wording around it is not. The paper's own sentence is the evidence quote for that row; on this page it is withheld with the article, and the article is at its publisher. Nothing inside quotation marks anywhere on this page is ours: if it is quoted, it is the paper's. Whose words a concept carries. A concept entry records a term the paper itself uses, so the term, the sense given for it, and its evidence are the paper's own words, quoted where the paper fixes them. This is the opposite of a sense entry in the tag table, which is the tagger's sentence about how the paper uses a term. Where a concept entry could not be grounded in the paper's words, the entry says so on its own row.
Tag layer © the Meta-Tagging Project, CC BY 4.0. The paper text is © its authors/publisher
and is linked, not reproduced. Excerpts shown are short quotations for identification and verification.
· back to index
· concepts · authors
· עברית
© 2026 Shir Sivroni · ORCID 0009-0000-2597-2824