You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Contrary to popular belief, an #instrumentalvariable can only provide BOUNDS for the effect. A point estimate (as in Mendelian randomization) requires additional unverifiable assumptions.
Our review of methods for bounds, led by Sonja Swanson, is just outhttps://www.tandfonline.com/doi/full/10.1080/01621459.2018.1434530 …
Code to implement the bounds methodology described in the paper is freely available in @SASsoftware here
https://www.hsph.harvard.edu/causal/software/ …
and here
https://github.com/CausalInference/IV-Bounds …
(R version currently under development)
¿Cómo dirías bounds en castellano? Acá educándome en la mañana.
Bound (English) = Cota (español/castellano) To non-Spanish speakers: language-wise, Castilian = Spanish. The choice of term depends on region (Argentinians say they speak “Castilian”) or context (Spaniards say they speak “Castilian” vs. “Catalan” but “Spanish” vs. “French”).
Obrigado :D
I tend to think that bounding is among the most underutilized approaches in epi causal inference. Scenarios where we can say that some effect is "at least as large as" are likely more clinically and policy relevant than associations with unknown net direction of bias.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
