augsynth_multiout: Fit Augmented SCM with multiple outcomes

In ebenmichael/augsynth: The Augmented Synthetic Control Method

Fit Augmented SCM with multiple outcomes

Description

Fit Augmented SCM with multiple outcomes

Usage

augsynth_multiout(
  form,
  unit,
  time,
  t_int,
  data,
  progfunc = c("Ridge", "None"),
  scm = T,
  fixedeff = FALSE,
  cov_agg = NULL,
  combine_method = "avg",
  ...
)

Arguments

form	outcome ~ treatment | auxillary covariates
unit	Name of unit column
time	Name of time column
t_int	Time of intervention
data	Panel data as dataframe
progfunc	What function to use to impute control outcomes Ridge=Ridge regression (allows for standard errors), None=No outcome model,
scm	Whether the SCM weighting function is used
fixedeff	Whether to include a unit fixed effect, default F
cov_agg	Covariate aggregation functions, if NULL then use mean with NAs omitted
combine_method	How to combine outcomes: ‘concat' concatenates outcomes and 'avg' averages them, default: ’avg'
...	optional arguments for outcome model

Value

augsynth object that contains:

• "weights"Ridge ASCM weights

• "l2_imbalance"Imbalance in pre-period outcomes, measured by the L2 norm

• "scaled_l2_imbalance"L2 imbalance scaled by L2 imbalance of uniform weights

• "mhat"Outcome model estimate

• "data"Panel data as matrices

ebenmichael/augsynth documentation built on March 20, 2024, 5:20 a.m.