Participe da nova comunidade do
Discord para discussões em tempo real, suporte de colegas e interação direta com a equipe do Meridian.
Premissas obrigatórias
Mantenha tudo organizado com as coleções
Salve e categorize o conteúdo com base nas suas preferências.
Em geral, não há um conceito de resultados potenciais na regressão, porque os modelos estimam as expectativas condicionais de uma variável de resposta. No entanto, de acordo com as principais premissas de troca condicional e consistência:
$$
E \Biggl(
\overset \sim Y_{g,t}^{
\left(\left\{
x_{g,t,i}^{(\ast)}
\right\}\right)
} \Big| \bigl\{z_{g,t,i}\bigr\}
\Biggr) = E \Biggl(
\overset \sim Y_{g,t} \Big|
\bigl\{z_{g,t,i}\bigr\}, \big\{x_{g,t,i}^{(\ast)}\bigr\} \Biggr)
$$
Principais premissas
Troca condicional:
\( \overset \sim Y_{g,t}^{(\{ x_{g,t,i}^{(\ast)} \})} \)é independente das variáveis aleatórias\(\bigl\{ X_{g,t,i}^{(\ast)} \bigr\}\) para qualquer cenário contrafactual\(\bigl\{ x_{g,t,i}^{(\ast)} \bigr\}\). Portanto, o conjunto de resultados potenciais é condicionalmente independente da decisão de execução de mídia histórica do anunciante.
Consistência:
\( \overset \sim Y_{g,t} = \overset \sim Y_{g,t}^{
(\{ x_{g,t,i}^{(\ast)} \})
} \) quando \(\bigl\{ X_{g,t,i}^{(\ast)} \bigr\} =
\bigl\{ x_{g,t,i}^{(\ast)} \bigr\}\). Portanto, o KPI observado do resultado potencial para o cenário contrafactual corresponde à execução de mídia histórica do anunciante.
Com base nessas premissas, você tem o resultado já declarado:
$$
E \Biggl( \overset \sim Y_{g,t}^{
\left(\left\{ x_{g,t,i}^{\ast} \right\}\right)
} \Big| \bigl\{ z_{g,t,i} \bigr\} \Biggr)
\overset{\text{exchangeability}}{=} E \Biggl( \overset \sim Y_{g,t}^{
\left(\left\{ x_{g,t,i}^{\ast} \right\}\right)
} \Big| \bigl\{ z_{g,t,i} \bigr\},\ \bigl\{ x_{g,t,i}^{(\ast)} \bigr\} \Biggr)
\overset{\text{consistency}}{=} E \Biggl( \overset \sim Y_{g,t}\ \Big|
\bigl\{ z_{g,t,i} \bigr\},\ \bigl\{ x_{g,t,i}^{(\ast)} \bigr\}
\Biggr)
$$
A premissa de consistência é bastante intuitiva e válida, a menos que o contrafactual esteja mal definido ou não seja representado com precisão nos dados.
Para mais informações, consulte Hernan MA, Robins JM, (2020) Causal Inference: What
If.
A premissa de troca condicional é um pouco menos intuitiva e só será válida se todas as variáveis confundidoras forem medidas e incluídas na matriz de controle \(\{z_{g,t,i}\}\). Variáveis confundidoras são tudo que tem um efeito causal no tratamento observado \(\{x_{g,t,i}\}\) e no resultado\(\{\overset \sim y_{g,t}\}\). No tratamento, isso significa um efeito do nível geral do orçamento do anunciante e da alocação entre canais, regiões geográficas ou períodos. Na prática, é difícil saber se todas as variáveis confundidoras são medidas, já que essa é apenas uma premissa, e não existe um teste estatístico que determine isso com base nos dados. No entanto, pode ser útil saber que a premissa de troca condicional será válida se você assumir um gráfico causal que atende a uma condição conhecida como o critério backdoor (Pearl, J., 2009). Para mais
informações, consulte Gráfico causal.
Exceto em caso de indicação contrária, o conteúdo desta página é licenciado de acordo com a Licença de atribuição 4.0 do Creative Commons, e as amostras de código são licenciadas de acordo com a Licença Apache 2.0. Para mais detalhes, consulte as políticas do site do Google Developers. Java é uma marca registrada da Oracle e/ou afiliadas.
Última atualização 2025-08-04 UTC.
[[["Fácil de entender","easyToUnderstand","thumb-up"],["Meu problema foi resolvido","solvedMyProblem","thumb-up"],["Outro","otherUp","thumb-up"]],[["Não contém as informações de que eu preciso","missingTheInformationINeed","thumb-down"],["Muito complicado / etapas demais","tooComplicatedTooManySteps","thumb-down"],["Desatualizado","outOfDate","thumb-down"],["Problema na tradução","translationIssue","thumb-down"],["Problema com as amostras / o código","samplesCodeIssue","thumb-down"],["Outro","otherDown","thumb-down"]],["Última atualização 2025-08-04 UTC."],[[["\u003cp\u003eRegression models can be used to estimate potential outcomes under the assumptions of conditional exchangeability and consistency.\u003c/p\u003e\n"],["\u003cp\u003eConditional exchangeability implies that potential outcomes are independent of historical media execution decisions, given confounding variables.\u003c/p\u003e\n"],["\u003cp\u003eConsistency means the observed outcome matches the potential outcome for the actual historical media execution.\u003c/p\u003e\n"],["\u003cp\u003eConfounding variables, which affect both treatment and outcome, must be measured and included for conditional exchangeability to hold.\u003c/p\u003e\n"],["\u003cp\u003eWhile there's no statistical test to guarantee conditional exchangeability, causal graphs and the backdoor criterion can help assess it.\u003c/p\u003e\n"]]],["Regression models typically lack potential outcomes, but under conditional exchangeability and consistency, we can derive a relevant result. Conditional exchangeability means potential outcomes are independent of historical media execution. Consistency dictates that observed outcomes match potential outcomes when treatment equals historical media execution. The key result is derived by first exchanging outcomes with potential outcomes, then aligning them with observed values under these assumptions. Conditional exchangeability relies on all confounders (variables affecting both treatment and outcome) being measured and can be assessed with causal graph analysis.\n"],null,["# Required assumptions\n\nGenerally speaking, there is no concept of potential outcomes in regression\nbecause regression models estimate conditional expectations of a response\nvariable. However, under the key assumptions of *conditional exchangeability*\nand *consistency*: \n$$ E \\\\Biggl( \\\\overset \\\\sim Y_{g,t}\\^{ \\\\left(\\\\left\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\right\\\\}\\\\right) } \\\\Big\\| \\\\bigl\\\\{z_{g,t,i}\\\\bigr\\\\} \\\\Biggr) = E \\\\Biggl( \\\\overset \\\\sim Y_{g,t} \\\\Big\\| \\\\bigl\\\\{z_{g,t,i}\\\\bigr\\\\}, \\\\big\\\\{x_{g,t,i}\\^{(\\\\ast)}\\\\bigr\\\\} \\\\Biggr) $$\n\n**Key assumptions**\n\n- Conditional exchangeability:\n\n \\\\( \\\\overset \\\\sim Y_{g,t}\\^{(\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\})} \\\\)\n is independent of the random variables\n \\\\(\\\\bigl\\\\{ X_{g,t,i}\\^{(\\\\ast)} \\\\bigr\\\\}\\\\) for any counterfactual scenario\n \\\\(\\\\bigl\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\bigr\\\\}\\\\). So, the set of potential outcomes\n is conditionally independent of the advertiser's historical media execution\n decision.\n- Consistency:\n\n \\\\( \\\\overset \\\\sim Y_{g,t} = \\\\overset \\\\sim Y_{g,t}\\^{\n (\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\})\n } \\\\) when \\\\(\\\\bigl\\\\{ X_{g,t,i}\\^{(\\\\ast)} \\\\bigr\\\\} =\n \\\\bigl\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\bigr\\\\}\\\\). So, the observed KPI realization of\n the potential outcome for the counterfactual scenario matching the\n advertiser's historical media execution.\n\nUnder these assumptions, you have the previously stated result: \n$$ E \\\\Biggl( \\\\overset \\\\sim Y_{g,t}\\^{ \\\\left(\\\\left\\\\{ x_{g,t,i}\\^{\\\\ast} \\\\right\\\\}\\\\right) } \\\\Big\\| \\\\bigl\\\\{ z_{g,t,i} \\\\bigr\\\\} \\\\Biggr) \\\\overset{\\\\text{exchangeability}}{=} E \\\\Biggl( \\\\overset \\\\sim Y_{g,t}\\^{ \\\\left(\\\\left\\\\{ x_{g,t,i}\\^{\\\\ast} \\\\right\\\\}\\\\right) } \\\\Big\\| \\\\bigl\\\\{ z_{g,t,i} \\\\bigr\\\\},\\\\ \\\\bigl\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\bigr\\\\} \\\\Biggr) \\\\overset{\\\\text{consistency}}{=} E \\\\Biggl( \\\\overset \\\\sim Y_{g,t}\\\\ \\\\Big\\| \\\\bigl\\\\{ z_{g,t,i} \\\\bigr\\\\},\\\\ \\\\bigl\\\\{ x_{g,t,i}\\^{(\\\\ast)} \\\\bigr\\\\} \\\\Biggr) $$\n\nThe consistency assumption is fairly intuitive, and holds unless the\ncounterfactual is poorly defined or is not accurately represented in the data.\nFor more information, see [Hernan MA, Robins JM, (2020) Causal Inference: What\nIf](https://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/).\n\nThe conditional exchangeability assumption is a bit less intuitive. This\nassumption holds if all confounding variables are measured and included in the\ncontrol array \\\\(\\\\{z_{g,t,i}\\\\}\\\\). *Confounding variables* are anything that has\na causal effect on both the observed treatment \\\\(\\\\{x_{g,t,i}\\\\}\\\\) and outcome\n\\\\(\\\\{\\\\overset \\\\sim y_{g,t}\\\\}\\\\). A causal effect on treatment can mean an effect\nof the advertiser's overall budget level, the allocation across channels, the\nallocation across geos, or the allocation across time periods. In practice, it\nis difficult to know whether all of the confounding variables are measured\nbecause it is purely an assumption, and there is no statistical test to\ndetermine this from your data. However, it can be helpful to know that the\nconditional exchangeability assumption holds if you assume a causal graph that\nmeets a condition known as the *backdoor criterion* (Pearl, J., 2009). For more\ninformation, see [Causal graph](/meridian/docs/basics/causal-graph)."]]