Skip to contents

Returns a tibble with all combos of predictors

Source: R/interpolate_newdata.R
interpolate_newdata.Rd

[Experimental]

Usage

interpolate_newdata(
  fit,
  by = NULL,
  x_values = NULL,
  at = NULL,
  arma = NULL,
  prior = FALSE
)

Arguments

fit

An mcpfit object.

by

Character vector of categorical or group-level columns to evaluate separately. Categorical model predictors are always evaluated separately.

x_values

Numeric vector of x-values to evaluate at.

at

Named list setting additional continuous predictors to fixed values. They default to their observed means. Family response auxiliaries can also be supplied as explicit scalar design values; e.g., at = list(N = 20).

arma

How AR/MA terms will be evaluated (see predict.mcpfit()). Defaults to TRUE when the fit includes ar() or ma() terms.

  • TRUE: Preserve the observed rows and response history for conditional AR/MA evaluation.

  • FALSE: Interpolate unconditional trends.

  • "replicate": Preserve the observed rows, because AR/MA terms depend on the previous rows.

prior

Logical. Increase the resolution around the change points of the prior draws (TRUE) instead of the posterior draws (FALSE, default).

Value

A tibble with

  • A column for the change-point variable (par_x)

  • All combinations of categorical predictor levels

  • Fixed values for additional continuous predictors

Details

This function synthesizes predictors for all combinations of predictor values. It is used internally in plot.mcpfit() and may be useful if you want to build your own custom plot.

The par_x variable will be interpolated with higher resolution around the change points where the values can change abruptly, but lower resolution in between to speed up the computation.

Categorical variables and requested grouping factors are combined factorially (all level combinations). Additional continuous predictors are held at their observed means, or at values supplied through at. Family-specific response auxiliaries are not interpolated. Supply binomial trial counts as a scalar in at or use newdata for a varying design.

Likelihood weights are needed only when evaluating log_lik() and therefore need not be supplied here.

Author

Jonas Kristoffer Lindeløv jonas@lindeloev.dk

Examples

# \donttest{
# Get predictors for a fit
newdata = interpolate_newdata(demo_fit)

# Fit summary
head(fitted(demo_fit, newdata))
#> # A tibble: 6 × 5
#>    time .mean   .sd .lower .upper
#>   <dbl> <dbl> <dbl>  <dbl>  <dbl>
#> 1 0.619  9.97 0.665   8.68   11.2
#> 2 1.62   9.97 0.665   8.68   11.2
#> 3 2.62   9.97 0.665   8.68   11.2
#> 4 3.62   9.97 0.665   8.68   11.2
#> 5 4.62   9.97 0.665   8.68   11.2
#> 6 5.62   9.97 0.665   8.68   11.2

# Predictions for each draw
prediction = predict(demo_fit, newdata, summary = FALSE)
head(prediction)
#> # A tibble: 6 × 13
#>    time  .row .chain .iteration .draw  cp_1  cp_2 Intercept_1 time_2 Intercept_3  time_3 sigma_1 .prediction
#>   <dbl> <int>  <int>      <int> <int> <dbl> <dbl>       <dbl>  <dbl>       <dbl>   <dbl>   <dbl>       <dbl>
#> 1 0.619     1      1          1     1  31.9  71.0        8.95  0.584        13.9 -0.155     3.48       11.0 
#> 2 0.619     1      1          2     2  31.7  71.4        9.87  0.566        15.8 -0.335     3.47        6.17
#> 3 0.619     1      1          3     3  32.1  70.5       11.4   0.526        12.6 -0.0956    4.02       15.4 
#> 4 0.619     1      1          4     4  33.3  72.8       10.1   0.561        13.7 -0.174     3.98        1.88
#> 5 0.619     1      1          5     5  31.7  69.9        9.97  0.466        13.8 -0.158     4.65       18.1 
#> 6 0.619     1      1          6     6  33.3  69.9       10.0   0.536        12.5 -0.140     3.98        4.23

# Custom plot
library(ggplot2)
plotdata = fitted(demo_fit, newdata)  # `width = .95` by default 
ggplot(plotdata, aes(x = time, y = .mean)) +
  geom_ribbon(aes(ymin = .lower, ymax = .upper), alpha = 0.3) +
  geom_line(lwd = 2) +
  geom_point(aes(y = response), data = demo_fit$data)

# }