plot_me: Visualizing Marginal Effects

In andreizhirnov/DAME: DAME: data-driven interpretation of interactive nonlinear models

Visualizing Marginal Effects

Description

plot_me produces a heatmap of the marginal effects of variable x plotted against the combinations of x and over and adds a scatterplot representing the joint distribution these two variables in the given sample. The size of the markers represents the number of observations.

Usage

plot_me(
  x,
  over,
  model = NULL,
  data = NULL,
  link = NULL,
  formula = NULL,
  coefficients = NULL,
  vcov = NULL,
  discrete = FALSE,
  discrete_step = 1,
  at = NULL,
  mc = FALSE,
  iter = 1000,
  heatmap_dim = c(100, 100),
  p = 0.05,
  weights = NULL
)

Arguments

x	a character string representing the name of the main variable of interest. Marginal effects will be computed for this variable.
over	a character string representing the name of the conditionning variable. DAME will be computed for the bins long the range of this variable.
model	fitted model object. The package works best with GLM objects and will extract the formula, dataset, family, coefficients, and the QR components of the design matrix if arguments formula, data, link, coefficients, and/or vcov are not explicitly specified.
data	the dataset to be used to compute marginal effects (if not specified, it is extracted from the fitted model object).
link	the name of the link function used in estimation (if not specified, it is extracted from the fitted model object).
formula	the formula used in estimation (if not specified, it is extracted from the fitted model object).
coefficients	the named vector of coefficients produced during the estimation (if not specified, it is extracted from the fitted model object).
vcov	the variance-covariance matrix to be used for computing standard errors (if not specified, it is extracted from the fitted model object).
discrete	logical. If TRUE, the function will compute the effect of a discrete change in x. If FALSE, the function will compute the partial derivative of x.
discrete_step	The size of a discrete change in x used in computations (used only if discrete=TRUE).
at	an optional named list of values of independent variables. These variables will be set to these value before computations. The remaining numeric variables (except x and over) will be set to their means. The remaining factor variables will be set to their modes.
mc	logical. If TRUE, the standard errors and confidence intervals will be computed using simulations. If FALSE (default), the delta method will be used.
iter	the number of interations used in Monte-Carlo simulations. Default = 1,000.
heatmap_dim	a numeric vector containing the number of rows and columns used for drawing the heatmap. Default = 100 each.
p	the singificance level for the marginal effects. Default = 0.05.
weights	an optional vector of sampling weights.

Details

plot_me provides a convenient way to interpret two-way interactions using heatmaps. It returns a ggplot object, which allows users to customize the plot using functions and layers from the ggplot2 package.

Author(s)

plot_me visualizes ME procedure described in Zhirnov, Moral, and Sedashov (2021) using the tools from ggplot2 package.

References

Zhirnov, Andrei, Mert Moral, and Evgeny Sedashov (2021). “Taking Distributions Seriously: On the Interpretation of the Estimates of Interactive Nonlinear Models.” Working paper.

Examples

##Poisson regression with 2 variables and interaction between them
## Not run: 
data <- data.frame(y = rpois(10000, 10), x2 = rpois(10000, 5), x1 = rpois(10000, 3))
y <- glm(y ~ x1 + x2 + x1*x2, data = data, family = "poisson")
## A contour-plot with 4 areas
library(ggplot2)
plot_me(model = y, data = data, x = "x1", over = "x2") +
    scale_fill_steps(low="yellow", high="red", n.breaks=4)
## A heatmap with smooth transition of colors
plot_me(model = y, data = data, x = "x1", over = "x2") +
    scale_fill_gradient(low="yellow", high="red")
## A heatmap with histograms at the edges
library(ggExtra)
g <- plot_me(model = y, data = data, x = "x1", over = "x2")
gt <- g + theme(legend.position="left") + scale_fill_gradient(low="yellow", high="red")
ggExtra::ggMarginal(gt, type="histogram", data=data, x=z, y=x)
## if more control over the histograms needed:
nbins <- sapply(data[c("x1","x2")], grDevices::nclass.FD)
ggExtra::ggMarginal(gt, type="histogram", data=data, x=z, y=x,
   xparams=list(bins=nbins['x2']), yparams=list(bins=nbins['x1']))

## End(Not run)
## Not run: 
## logit
m <- glm(any_dispute ~ flows.ln*polity2 + gdp_pc, data=strikes, family="binomial")
summary(m)
plot_me(model = m, x = "flows.ln", over = "polity2") +
    scale_fill_gradient(low="yellow", high="red") +
    labs(x="Polity", y="ln(FDI flows)")

## End(Not run)

andreizhirnov/DAME documentation built on Nov. 21, 2022, 5:55 p.m.