plot_placebos {tidysynth}

plot_placebos

Description

Plot the difference between the observed and sythetic control unit for the treated and the placebo units. The difference captures the causal quantity (i.e. the magnitude of the difference between the observed and counterfactual case). Plotting the actual treated observation against the placebos captures the likelihood (or rarity) of the observed differenced trend.

Usage

plot_placebos(data, time_window = NULL, prune = TRUE)

Arguments

data

nested data of type tbl_df.

time_window

time window of the tbl_df plot.

prune

boolean flag; if TRUE, then all placebo cases with a pre-period RMSPE exceeding two times the treated unit pre-period RMSPE are pruned; Default is TRUE.

Details

The function provides a pruning rule where all placebo cases with a pre-period root mean squared predictive error (RMSPE) exceeding two times the treated unit pre-period RMSPE are pruned. This helps overcome scale issues when a particular placebo case has poor fit in the pre-period.

See documentation on ?synthetic_control on how to generate placebo cases. When initializing a synth pipeline, set the generate_placebos argument to TRUE. The processing pipeline remains the same.

Value

ggplot object of the difference between the observed and synthetic trends for the treated and placebo units.

Examples

# Smoking example data
data(smoking)

smoking_out <-
smoking %>%

# initial the synthetic control object
synthetic_control(outcome = cigsale,
                  unit = state,
                  time = year,
                  i_unit = "California",
                  i_time = 1988,
                  generate_placebos=TRUE) %>%

# Generate the aggregate predictors used to generate the weights
  generate_predictor(time_window=1980:1988,
                     lnincome = mean(lnincome, na.rm = TRUE),
                     retprice = mean(retprice, na.rm = TRUE),
                     age15to24 = mean(age15to24, na.rm = TRUE)) %>%

  generate_predictor(time_window=1984:1988,
                     beer = mean(beer, na.rm = TRUE)) %>%

  generate_predictor(time_window=1975,
                     cigsale_1975 = cigsale) %>%

  generate_predictor(time_window=1980,
                     cigsale_1980 = cigsale) %>%

  generate_predictor(time_window=1988,
                     cigsale_1988 = cigsale) %>%


  # Generate the fitted weights for the synthetic control
  generate_weights(optimization_window =1970:1988,
                   Margin.ipop=.02,Sigf.ipop=7,Bound.ipop=6) %>%

  # Generate the synthetic control
  generate_control()

# Plot the observed and synthetic trend
smoking_out %>% plot_placebos(time_window = 1970:2000)