{"paper":{"title":"Estimator Averaging of Local Projection and VAR Impulse Responses","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Estimator averaging optimally combines local projections and VARs to minimize mean squared error of impulse responses at each horizon.","cross_cats":[],"primary_cat":"econ.EM","authors_text":"Balazs Vonnak, Chaoyi Chen, Elena Pesavento","submitted_at":"2026-05-06T21:24:39Z","abstract_excerpt":"Local projections (LP) and vector autoregressions (VAR) are the two standard tools for impulse response analysis, but they often display a finite-sample trade-off: LP is typically less biased but more volatile, while VAR is more precise but can be biased under misspecification. We propose an easy-to-implement estimator-averaging approach that combines LP and VAR at each horizon by minimizing the mean squared error of the impulse response itself, rather than in-sample fit. We derive closed-form oracle weights for this finite-sample risk problem, develop feasible AR-sieve-bootstrap procedures, a"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We propose an easy-to-implement estimator-averaging approach that combines LP and VAR at each horizon by minimizing the mean squared error of the impulse response itself, rather than in-sample fit. We derive closed-form oracle weights for this finite-sample risk problem, develop feasible AR-sieve-bootstrap procedures...","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"For a benchmark class of short-memory linear data generating processes in which LP and VAR are both consistent, we establish the consistency and limiting distribution of the feasible averaged estimator.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A new estimator-averaging method combines LP and VAR impulse responses by minimizing MSE at each horizon, with closed-form weights and bootstrap procedures, showing risk reduction in Monte Carlo and empirical tests.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Estimator averaging optimally combines local projections and VARs to minimize mean squared error of impulse responses at each horizon.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e53beffedc1eb8c97981d0761781c65d4e8243103afab4a65e7f4c4b80e7b80d"},"source":{"id":"2605.05456","kind":"arxiv","version":2},"verdict":{"id":"cf5224ba-39a4-4eda-b5b9-bbea592bab76","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T15:23:04.582563Z","strongest_claim":"We propose an easy-to-implement estimator-averaging approach that combines LP and VAR at each horizon by minimizing the mean squared error of the impulse response itself, rather than in-sample fit. We derive closed-form oracle weights for this finite-sample risk problem, develop feasible AR-sieve-bootstrap procedures...","one_line_summary":"A new estimator-averaging method combines LP and VAR impulse responses by minimizing MSE at each horizon, with closed-form weights and bootstrap procedures, showing risk reduction in Monte Carlo and empirical tests.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"For a benchmark class of short-memory linear data generating processes in which LP and VAR are both consistent, we establish the consistency and limiting distribution of the feasible averaged estimator.","pith_extraction_headline":"Estimator averaging optimally combines local projections and VARs to minimize mean squared error of impulse responses at each horizon."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.05456/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T09:41:11.359395Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T20:31:19.658387Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T13:33:55.228515Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"82c3e57d7211b2cd72638cfa22986c84f4b89257d7d897af7a3402aaf29fe53f"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}