# addhlines: Add horizontal lines to plot. In gamclass: Functions and Data for a Course on Modern Regression and Classification

## Description

This is designed for adding horizontal lines that show predicted values to a plot of observed values versus x-values, in `rpart` regression. Where predicted values change between two successive x-values lines are extended to the midway point. This reflects the way that `predict.rpart` handles predictions for new data.

## Usage

 `1` ```addhlines(x, y, ...) ```

## Arguments

 `x` Vector of predictor variable values. `y` Vector of predicted values. `...` Additional graphics parameters, for passing through to the `lines()` function.

## Value

Lines are added to the current graph.

John Maindonald

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24``` ```x <- c(34, 18, 45, 18, 27, 24, 34, 20, 24, 28, 21, 18) y <- c(14, 11, 12, 9, 4, 11, 6, 9, 4, 10, 9, 2) hat <- c(10.5, 7.75, 10.5, 7.75, 7, 7, 10.5, 7.75, 7, 10.5, 7, 7.75) plot(x, y) addhlines(x, hat, lwd=2, col="gray") ## The function is currently defined as function(x,y, ...){ ordx <- order(x) xo <- x[ordx] yo <- y[ordx] breaks <- diff(yo)!=0 xh <- c(xo,0.5*(xo[c(FALSE,breaks)]+xo[c(breaks, FALSE)])) yh <- yo[c(TRUE, breaks)] y3 <- x3 <- numeric(3*length(xh)-1) loc1 <- seq(from=1, to=length(x3), by=3) x3[loc1] <- xh x3[loc1+1]<- c(xh[-1], max(x)) x3[loc1[-length(loc1)]+2] <- NA y3[loc1[-length(loc1)]+2] <- NA y3[loc1] <- yh y3[loc1+1] <- yh lines(x3,y3, ...) } ```

gamclass documentation built on Nov. 14, 2020, 1:07 a.m.