obs_v_pred_plot: Observed against predicted plot

In sujit-sahu/bmstdr: Bayesian Modeling of Spatio-Temporal Data with R

Observed against predicted plot

Description

Observed against predicted plot

Usage

obs_v_pred_plot(
  yobs,
  predsums,
  segments = TRUE,
  summarystat = "median",
  plotit = TRUE
)

Arguments

yobs	A vector containing the actual observations
predsums	A data frame containing predictive summary statistics with the same number of rows as the length of the vector yobs. The data frame must have columns named as meanpred, medianpred, sd, low and up. Ideally this argument should be the output of the command get_validation_summaries.
segments	Logical: whether to draw line segments for the prediction intervals.
summarystat	Can take one of two values "median" (default) or "mean" indicating which one to use for the plot.
plotit	Logical scalar value: whether to plot the predictions against the observed values.

Value

Draws a plot only after removing the missing observations. It also returns a list of two ggplot2 objects: (i) a plot with intervals drawn pwithseg and (ii) a plot without the segments drawn: pwithoutseg and (iii) a simple plot not showing the range of the prediction intervals.

Examples

set.seed(4)
vrows <- sample(nrow(nysptime), 100)
M1 <- Bsptime(model="lm", formula=y8hrmax~xmaxtemp+xwdsp+xrh, data=nysptime, 
validrows=vrows, scale.transform = "SQRT")
psums <-  get_validation_summaries(M1$valpreds)
oplots <- obs_v_pred_plot(yobs=M1$yobs_preds$y8hrmax, predsum=psums)
names(oplots)
plot(oplots$pwithoutseg)
plot(oplots$pwithseg)

sujit-sahu/bmstdr documentation built on Jan. 30, 2024, 1:40 p.m.