Pular para o conteúdo

Tarefa paper_repro.mmv_2011.tab20_1_c2.tool

paper_reprotoolT1publicépoca 1, tentativa 1Execução: 20261015T165911Z-suite-paper_repro-baseline-8b4f85

Veredito

passouPassou em todas as verificações obrigatórias.
Calculado pelo agente
effect =0,03348180.03348175396085768
Jump in Support at the cutoff of Income_Below
via infer.rdd (call_96f91092)
Valor esperado (gabarito)
effect =0,03348180.03348175396085784
statsmodels 0.15.0: the reference specification of mmv_2011 tab20_1_c2 (infer.rdd, figure 'effect') on an independent code path
Como foi validado
3 de 3 verificações obrigatórias aprovadas
relativa 0.000001 · absoluta 4.419589592144583e-8

Pergunta feita ao analista

Tarefa de ferramenta: nenhuma pergunta; a especificação de referência é chamada direto.

Como o agente resolveu e como foi validado

Resumo passo a passo do que o agente executou, por que o caminho está correto e como cada verificação conferiu o resultado.

  1. 1Preparação da base de dados
    gov_transfers— Households near the income cutoff of Uruguay's PANES cash-transfer programme, the extract of the survey of Manacorda, Miguel and Vigorito (2011), 'Government Transfers and Political Support', distributed with the causaldata R package and used in chapter 20 of Huntington-Klein's 'The Effect': one row per household (1,948 rows), only households whose income measure is within 0.02 of the cutoff. Households below the cutoff were eligible for the transfers. Columns: rownames, the row number; Income_Centered, the household's income measure minus the eligibility cutoff (negative values are below the cutoff, hence eligible); Income_Below, minus Income_Centered, added by the task author: how far below the cutoff the household's income measure is (positive values are eligible); Education, average years of education of household members aged 16 and older (missing for some households); Age, average age of household members; Participation, 1 if the household participated in the transfer programme; Support, the household's support for the government (0, 0.5 or 1; higher means more support).
    Transformação "derive" aplicada à base "gov_transfers" (1948 → 1948 linhas). {"args":{"column":"Income_Centered"},"name":"Income_Below","op":"negate"}
  2. 2Chamada que gerou a resposta finalinfer.rdd(Regressão descontínua (RDD))citada na respostacoincide com a referência
    Ver chamada call_96f91092 nos detalhes
    Argumentos principais
    bandwidth:
    0,010.01
    cutoff:
    00
    dataset:
    "gov_transfers"
    kernel:
    "triangular"
    level:
    0,950.95
    polynomial:
    11
    running:
    "Income_Below"
    y:
    "Support"
    Resultado calculado
    • effect = 0,0334818 0.03348175396085768
      Jump in Support at the cutoff of Income_Below
    Diagnósticos: bandwidth = 0,010.01 (half-width of the window around the cutoff (informada)) · bandwidth_method = "user" · n_left = 400400 (observations with positive weight below the cutoff) · n_right = 537537 (observations with positive weight at or above the cutoff) · kernel = "triangular" · polynomial = 11 · bandwidth_sensitive = false (the estimates at half and double the bandwidth fall inside the main interval) · n_obs = 1.9481948 (rows used to choose the bandwidth) · se_type = "HC1" (heteroskedasticity-robust standard errors (HC1))
  3. 3Conferência de lastro e resposta final
    • #0figura próximaeffect
      software reference
    • #1figura próximaeffect
      software reference, SE
    • #2figura próximaeffect
      published cell
    • #3figura próximaeffect
      published cell, SE (known deviation: unverified)

Aprofundamento técnico para especialistas

Acesso direto às fontes primárias, tabela formal de verificações, comparação de argumentos, trajetória completa com prompts brutos e somas SHA-256.

Valores esperados e de onde vêm

mmv_2011.tab20_1_c2.software

Fonte
statsmodels 0.15.0: the reference specification of mmv_2011 tab20_1_c2 (infer.rdd, figure 'effect') on an independent code path
Tipo
software
Localização
mmv_2011: Table 20.1 (The Effect of Government Payments on Support, Regression Discontinuity Estimates), column 'Linear with Kernel Weight', row 'Participation'
Valor
0,03348180.03348175396085784 (se 0,04419590.04419589592144583)
Conferido por
execução de software · sm.WLS(weights=1 - |d|/h)(Support, [const, treat, d·(1-treat), d·treat]).fit(cov_type='HC1') with d = Income_Below - 0, rows |d| <= 0.01, treat = d >= 0 on gov_transfers (the file pinned at sha256 51312ecd5b3cb22b5dc027966a81e0a8579f24a708ce4b476674fc527652ebd0, after its 1 declared transforms); n_obs = 1948, n_left = 400, n_right = 537. The tool infer.rdd (magi_infer, numpy) does not call statsmodels. Note: the strict rule (|d| < h, treated when d > 0) selects the same rows and treatment. Written by scripts/paper_repro_reference.py.
Implementação
independente da ferramenta
Endereço
https://pypi.org/project/statsmodels/0.15.0/
  • #0 figura próxima: 0,03348180.03348175396085784 · relativa 0.000001 · absoluta 4.419589592144583e-8
  • #1 figura próxima: 0,04419590.04419589592144583 · relativa 0.000001 · absoluta 4.419589592144583e-8

mmv_2011.tab20_1_c2.paper

Fonte
Huntington-Klein, N. (2021). The Effect: An Introduction to Research Design and Causality. CRC Press; online edition, chapter 20 (Regression Discontinuity), replicating Manacorda, Miguel and Vigorito (2011), AEJ: Applied 3(3): 1-28.
Tipo
textbook
Localização
Table 20.1 (The Effect of Government Payments on Support, Regression Discontinuity Estimates), column 'Linear with Kernel Weight', row 'Participation'
Página
online edition, chapter 20, Table 20.1
Trecho
Participation 0.033 (0.021)
Valor como publicado
0.033
Erro-padrão como publicado
0.021
Valor
0,0330.033 (se 0,0210.021) support for the government (0, 0.5, 1)
Conferido por
fonte primária · Read on 2026-10-15 from the free online edition of The Effect (theeffectbook.net, chapter 20); read_copy_sha256 is the sha256 of the chapter's decompressed HTML as served that day. Table 20.1 is a modelsummary table of the R code printed above it: Support on Income_Centered*Participation plus I(Income_Centered^2)*Participation (column 'Quadratic', all 1,948 households) and Support on Income_Centered*Participation weighted by max(0, 1 - 100|Income_Centered|) (column 'Linear with Kernel Weight', Num.Obs. 937); the table's last rows state 'Std.Errors HC3 HC3'. Data check made before any estimate: Participation equals (Income_Centered <= 0) on every row, no row has Income_Centered exactly 0 or exactly 0.01 in absolute value, every |Income_Centered| is below 0.02, and 937 rows lie strictly within 0.01 of the cutoff (400 above it, 537 below), as the table's Num.Obs.
sha256 da cópia lida
b6f751cfa056dc32718bdd79f6b25518a6acd9e9923a8fd12ef29ac9342413eb (identifica a cópia lida, não um arquivo canônico)
Desvio conhecido
não conferido
Recorded before this track's software-reference run and before any dry run, from the printed table and the research probe: Table 20.1 states that its standard errors are HC3, while infer.rdd computes HC1 (its only unclustered option); the research note (research-papers.md section 3.15, 2026-10-14) already reported an HC1 error near 0.044 for this specification against the printed 0.021. The printed cell is recorded and non-required on T1. Corrected at the B-join review: the HC type does not explain the gap. The column's printed coefficients match exactly, while the HC0, HC1, HC3 and classical errors of this weighted fit (0.037 to 0.0445) all differ from the printed 0.021; the convention behind the printed SE is not documented, so the kind is unverified (SE only; it was se_convention).
Endereço
https://theeffectbook.net/ch-RegressionDiscontinuity.html
Licença
Copyright Nick Huntington-Klein (The Effect, CRC Press); cited, not reproduced.
  • #2 figura próxima: 0,0330.033 · published_cell d=3 · absoluta 0.000500001
  • #3 figura próxima: 0,0210.021 · published_cell d=3 (the SE as printed) · absoluta 0.000500001
Especificação técnica do gabarito

Manacorda, M., Miguel, E. and Vigorito, A. (2011). Government Transfers and Political Support. American Economic Journal: Applied Economics 3(3): 1-28 (doi:10.1257/app.3.3.1); the cells are those of its replication in Huntington-Klein, N. (2021), The Effect: An Introduction to Research Design and Causality, chapter 20 (Regression Discontinuity), Table 20.1. Table 20.1 (The Effect of Government Payments on Support, Regression Discontinuity Estimates), column 'Linear with Kernel Weight', row 'Participation' (source mmv_2011.tab20_1_c2.paper); software reference mmv_2011.tab20_1_c2.software: statsmodels 0.15.0: the reference specification of mmv_2011 tab20_1_c2 (infer.rdd, figure 'effect') on an independent code path

Verificações

Cada verificação, o que esperava, o que observou e a chamada que leu
#VerificaçãoObrigatóriaResultadoEsperadoObservadoTolerânciaFonteChamada
0
figura próximaeffect
software reference
0.03348175396 vs 0.03348175396085784
obrigatóriapassou0,03348180.033481753960857840,03348180.03348175396085768relativa 0.000001 · absoluta 4.419589592144583e-8mmv_2011.tab20_1_c2.software—
1
figura próximaeffect
software reference, SE
0.04419589592 vs 0.04419589592144583
obrigatóriapassou0,04419590.044195895921445830,04419590.04419589592144589relativa 0.000001 · absoluta 4.419589592144583e-8mmv_2011.tab20_1_c2.software—
2
figura próximaeffect
published cell
0.03348175396 vs 0.033
obrigatóriapassou0,0330.0330,03348180.03348175396085768published_cell d=3 · absoluta 0.000500001mmv_2011.tab20_1_c2.paper—
3
figura próximaeffect
published cell, SE (known deviation: unverified)
0.04419589592 vs 0.021
registradafalhou0,0210.0210,04419590.04419589592144589published_cell d=3 (the SE as printed) · absoluta 0.000500001mmv_2011.tab20_1_c2.paper—

Caminho de referência e caminho do agente

A chamada de referência ao lado da chamada que deu a resposta, argumento por argumento.

a própria especificação da tarefa

Referência: infer.rddChamada do agente: infer.rdd (call_96f91092)igual
Argumentos da referência e do agente
ArgumentoReferênciaAgente
bandwidth0,010.010,010.01igual
cluster——não fixado pela referência
cutoff0000igual
dataset"gov_transfers""gov_transfers"igual
kernel"triangular""triangular"igual
level—0,950.95não fixado pela referência
polynomial1111igual
running"Income_Below""Income_Below"igual
subset——não fixado pela referência
y"Support""Support"igual

Trajetória

Cada evento na ordem em que aconteceu: turnos do modelo, chamadas de ferramentas, guardas, transformações e pontuação.

  1. 0Início

    Tarefa de ferramenta: nenhuma pergunta; a especificação de referência é chamada direto.

    gov_transfers (agent)

    Households near the income cutoff of Uruguay's PANES cash-transfer programme, the extract of the survey of Manacorda, Miguel and Vigorito (2011), 'Government Transfers and Political Support', distributed with the causaldata R package and used in chapter 20 of Huntington-Klein's 'The Effect': one row per household (1,948 rows), only households whose income measure is within 0.02 of the cutoff. Households below the cutoff were eligible for the transfers. Columns: rownames, the row number; Income_Centered, the household's income measure minus the eligibility cutoff (negative values are below the cutoff, hence eligible); Income_Below, minus Income_Centered, added by the task author: how far below the cutoff the household's income measure is (positive values are eligible); Education, average years of education of household members aged 16 and older (missing for some households); Age, average age of household members; Participation, 1 if the household participated in the transfer programme; Support, the household's support for the government (0, 0.5 or 1; higher means more support).

    file 51312ecd5b3cb22b5dc027966a81e0a8579f24a708ce4b476674fc527652ebd0 · frame —
    1. derive {"args":{"column":"Income_Centered"},"name":"Income_Below","op":"negate"}
  2. 1Transformação
    gov_transfers · derive {"args":{"column":"Income_Centered"},"name":"Income_Below","op":"negate"}
    1948 → 1948 linhas
    Frame antes → depois: 2600b9da8866 → ee25bc867239
  3. 2Ferramenta
    infer.rdd— Regressão descontínua (RDD)v0.1.0call_96f91092decidido pelo códigoreexecutável
    Argumentos
    {
      "bandwidth": 0.01,
      "cluster": null,
      "cutoff": 0,
      "dataset": "gov_transfers",
      "kernel": "triangular",
      "level": 0.95,
      "polynomial": 1,
      "running": "Income_Below",
      "subset": null,
      "y": "Support"
    }
    Figuras call_96f91092
    ChaveValorValor brutoErro-padrãoIC
    effect
    Jump in Support at the cutoff of Income_Below
    0,03348180.033481753960857680,04419590.04419589592144589-0,0531406 – 0,120104
    Diagnósticos
    • bandwidth = 0,010.01 (half-width of the window around the cutoff (informada))
    • bandwidth_method = "user"
    • n_left = 400400 (observations with positive weight below the cutoff)
    • n_right = 537537 (observations with positive weight at or above the cutoff)
    • kernel = "triangular"
    • polynomial = 11
    • bandwidth_sensitive = false (the estimates at half and double the bandwidth fall inside the main interval)
    • n_obs = 1.9481948 (rows used to choose the bandwidth)
    • se_type = "HC1" (heteroskedasticity-robust standard errors (HC1))
    Derivação matemática e pressupostos
    effect
    1. Ajuste linear ponderado (kernel triangular) de cada lado do limiar, com janela h (informada); o efeito é a diferença entre os dois ajustes no limiar.→ 0,0334818 (0.03348175396085768)
    2. Erro-padrão robusto à heterocedasticidade (HC1): a raiz da variância estimada do coeficiente.ep = √V_jj→ 0,0441959 (0.04419589592144589)
    3. Intervalo de 95% de confiança: a estimativa mais ou menos o valor crítico da distribuição normal vezes o erro-padrão.IC = τ ± c · ep
    Pressupostos do estimador
    • units cannot precisely control the running variable around the cutoff
    • every other determinant of the outcome is continuous at the cutoff
    Tabelas (1)
    {
      "sensitivity": [
        {
          "bandwidth": 0.005,
          "ci_high": 0.11400899969691185,
          "ci_low": -0.16831958537112499,
          "estimate": -0.027155292837106566,
          "multiplier": 0.5,
          "n_left": 186,
          "n_right": 276,
          "se": 0.07202392168810465
        },
        {
          "bandwidth": 0.01,
          "ci_high": 0.12010411823137229,
          "ci_low": -0.05314061030965694,
          "estimate": 0.03348175396085768,
          "multiplier": 1,
          "n_left": 400,
          "n_right": 537,
          "se": 0.04419589592144589
        },
        {
          "bandwidth": 0.02,
          "ci_high": 0.15743353252374911,
          "ci_low": 0.0342718586238353,
          "estimate": 0.09585269557379221,
          "multiplier": 2,
          "n_left": 821,
          "n_right": 1127,
          "se": 0.03141937170055098
        }
      ]
    }
    Hashes SHA-256 e metadados de reprodução
    Digests dos dados
    gov_transfers ee25bc8672392d74aa124b1f295a152c7535bd2d306ed08a52b235e4ddbe5113
    sha256 do resultado
    dffa233e7e02748a8fdf865e5d9a70a8086da6d205a0e42d584459a7e2656f2a
  4. 3Pontuação
    #0figura próximapassouobrigatória0.03348175396 vs 0.03348175396085784
  5. 4Pontuação
    #1figura próximapassouobrigatória0.04419589592 vs 0.04419589592144583
  6. 5Pontuação
    #2figura próximapassouobrigatória0.03348175396 vs 0.033
  7. 6Pontuação
    #3figura próximafalhouregistrada0.04419589592 vs 0.021

Épocas × verificações

Resultado de cada verificação em cada época pontuada
#0123
Época 1passoupassoupassoufalhou

Tentativas

Cada tentativa é seu próprio arquivo e é publicada; só a última de cada época é pontuada.

Downloads desta tarefa

Pacote da tarefa (.zip)
50,5 KB · sha256 d88320321424941fbc53b42c9da88b081bafdc96fbe0b593398d5c51ddca3137

O hash de um zip é só de transporte: o que ancora a integridade é o RUN_SHA256SUMS que vai dentro dele, cópia exata do SHA256SUMS da execução.

Downloads desta tarefa
ArquivoTamanhosha256
sources/mmv_2011.tab20_1_c2.paper.json2,9 KB4fb38d0f878f500514322bdc3defabee0017164ac1a315d3f7dac239abe70ff4
sources/mmv_2011.tab20_1_c2.software.json1,4 KB8468ad5147c0b056840c6447f8a9a97c76fd843d855294f03517da1f7b685fe5
specs/paper_repro.mmv_2011.tab20_1_c2.tool.json6,1 KB97c69794d5c871c08e7f5f50a9d8c44063eedcac41b37e8403f2d0ba4aee5a6c
traces/paper_repro.mmv_2011.tab20_1_c2.tool.e1.a1.jsonl4,9 KB7698717d687d1a1640234c7aea851739347ad9fb734fbb0f3a73043bc262a8e2
trajectories/paper_repro.mmv_2011.tab20_1_c2.tool.e1.a1.jsonl8,8 KB7ec1d2b49277339e01de36a675d06d4c11fe598bb425bbbee75e6a8a476a4a27