An all-in-one DAG-driven robustness check. Generate publication-quality reports that classify variables by causal role, compare the significance of DAG-derived models, and explicitly target estimands.
DAGassist does:You can install DAGassist with:
#install.packages("DAGassist")
library(DAGassist) Or you can install the development version from GitHub with:
# install.packages("devtools")
devtools::install_github("grahamgoff/DAGassist")Simply provide a dagitty() object and a regression call
and DAGassist will create a report classifying variables by
causal role, and compare the specified regression to minimal and
canonical models.
DAGassist(dag = dag_model,
formula = lm(Y ~ X + M + C + Z + A + B, data = df),
estimand = c("total", "direct")
)
#> DAGassist Report:
#>
#> Roles:
#> variable role Exp. Out. conf med col dOut dMed dCol dConfOn dConfOff NCT NCO
#> X exposure x
#> Y outcome x
#> Z confounder x
#> M mediator x
#> C collider x x x
#> A nco x
#> B nco x
#>
#> (!) Bad controls in your formula: {M, C}
#> Minimal controls 1: {Z}
#> Canonical controls: {A, B, Z}
#>
#> Formulas:
#> original: Y ~ X + M + C + Z + A + B
#>
#> Balance diagnostics:
#> legend: (S)MD compares covariate means between the Original complete-case sample
#> and each spec's sample; |(S)MD| > 0.10 flags a covariate whose sample
#> composition shifts (binary vars use a raw difference in means).
#> Original vs Minimal 1: n = 2000 vs 2000 balanced
#> Original vs Canonical: n = 2000 vs 2000 balanced
#> Minimal 1 vs Canonical: n = 2000 vs 2000 balanced
#>
#> Model comparison:
#>
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | Original | Total Minimal 1 (Raw) | Total Canonical (Raw) | Total Minimal 1 (Weighted) | Total Canonical (Weighted) | Direct (Raw) | Direct (Weighted) |
#> +==========+===========+=======================+=======================+============================+============================+==============+===================+
#> | X | 0.452*** | 1.256*** | 1.256*** | 1.084*** | 1.097*** | 0.719*** | 0.620*** |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | (0.032) | (0.027) | (0.026) | (0.018) | (0.018) | (0.023) | (0.037) |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | M | 0.514*** | | | | | | |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | (0.021) | | | | | | |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | C | 0.343*** | | | | | | |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | (0.019) | | | | | | |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | Z | 0.249*** | 0.311*** | 0.309*** | | | 0.294*** | 0.440*** |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | (0.027) | (0.034) | (0.033) | | | (0.029) | (0.043) |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | A | 0.152*** | | 0.187*** | | | 0.180*** | 0.188*** |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | (0.021) | | (0.026) | | | (0.023) | (0.036) |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | B | -0.069*** | | -0.057* | | | -0.078*** | -0.099** |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | | (0.021) | | (0.026) | | | (0.023) | (0.038) |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | Num.Obs. | 2000 | 2000 | 2000 | 2000 | 2000 | 2000 | 2000 |
#> +----------+-----------+-----------------------+-----------------------+----------------------------+----------------------------+--------------+-------------------+
#> | R2 | 0.818 | 0.706 | 0.714 | 0.655 | 0.664 | | |
#> +==========+===========+=======================+=======================+============================+============================+==============+===================+
#> | + p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001 |
#> +==========+===========+=======================+=======================+============================+============================+==============+===================+
#>
#> Weight diagnostics:
#> legend: w range reports the min-max weights by group; ESS is kish effective sample size.
#> Total Minimal 1 (Weighted): w range=0.024..371.8 | ESS (weighted)=56.15 [LOW_ESS,EXTREME_W]
#> Total Canonical (Weighted): w range=0.02283..339.7 | ESS (weighted)=64.48 [LOW_ESS,EXTREME_W]
#>
#> Roles legend: Exp. = exposure; Out. = outcome; CON = confounder; MED = mediator; COL = collider; dOut = descendant of outcome; dMed = descendant of mediator; dCol = descendant of collider; dConfOn = descendant of a confounder on a back-door path; dConfOff = descendant of a confounder off a back-door path; NCT = neutral control on treatment; NCO = neutral control on outcomeOptionally, users can generate visual output via dotwhisker plots:
