plot.pred.int: Plot objects of class 'pred.int' for predicted interval plots

Description Usage Arguments Value Author(s) References See Also Examples

Description

Plot objects of class pred.int for predicted interval plots

Usage

1
2
3
## S3 method for class 'pred.int'
plot(x, conf.int = TRUE, vline = NA, which = NULL, axes = TRUE, 
    pi.col.fun=NULL, ci.col=2, main, xlab, ylab, xlim, ...)

Arguments

x

Object of class pred.int containing the PIPs data for plotting

conf.int

Print confidence interval for observed data? TRUE (default)/FALSE

vline

Vector of x-values for vertical lines to print on the graph. These may represent superiority/inferiority bounds or other x-values of interest

which

Only create graphs for some of the comparisons

axes

TRUE/FALSE. Print axes on graph? Default is TRUE. Probably you should only suppress axes now if you will add them later (for example if you don't like the default axes)

pi.col.fun

An optional one parameter function that takes a number between 0 and 100 and returns a color. This can be used to color the predicted intervals different colors. The input argument is the position of the interval on the vertical axis of the graph. Default coloring is three shades of gray: Percentiles (0-10) and (90-100) are light gray, (10-25) and (75-90) are darker, and (25-75) is darkest

ci.col

Color for effect estimate and confidence interval for the observed data. Default is 2 (second color in palette).

main

Main title of graph. If blank, a default will be used. Can be either a single title or a vector of titles. If a vector, the first will be used for the first graph, the second for the second graph, etc... If any title contains the string "#BY#", this will be replaced with the name of the comparison (i.e. "B vs A")

xlab

Label on xaxis. If blank, a default will be used.

ylab

Label on xaxis. If blank, a default will be used.

xlim

Limits of xaxis (as vector of length 2). Default is large enough to contain the predicted intervals and confidence interval. Limits narrower than the defaults will be ignored.

...

Other options will be passed through to the plot.default function.

Value

No return value. Called for its side effect.

Author(s)

Daniel G. Muenz, Ray Griner rgriner@sdac.harvard.edu, Huichao Chen, Lijuan Deng, Sachiko Miyahara, and Scott R. Evans evans@sdac.harvard.edu, with contributions from Lingling Li, Hajime Uno, and Laura M. Smeaton.

See package documentation for affiliations and contributions.

References

Evans SR, Li L, Wei LJ, "Data Monitoring in Clinical Trials Using Prediction", Drug Information Journal, 41:733-742, 2007.

Li L, Evans SR, Uno H, Wei LJ. "Predicted Interval Plots: A Graphical Tool for Data Monitoring in Clinical Trials", Statistics in Biopharmaceutical Research, 1:4:348-355, 2009.

See Also

PIPS

pred.int

print.pred.int

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# Make some fake data
myY<-c(rep(1,times=20),rep(0,times=80),rep(1,times=25),rep(0,times=25))
myGroup<-c(rep('A',100),rep('B',50))

# Run the programs
pips <- pred.int(y=myY, group=myGroup, N=c(200,100), 
                 data.type="binary", iters=100)
print(pips)
plot(pips)

# Run demo(package="PIPS") for more examples.

Example output

Sample sizes:
  Observed Simulated Total
A      100       100   200
B       50        50   100

Point estimates and 95% confidence intervals from observed data:
       Point Lower Bound Upper Bound
B vs A   0.3      0.1408      0.4592

Point estimates and 95% predicted intervals from observed+simulated data:
B vs A:
    Point Lower Bound Upper Bound
1   0.200     0.08751      0.3125
2   0.200     0.08751      0.3125
3   0.205     0.09354      0.3165
4   0.230     0.11685      0.3432
...                              
97  0.365     0.25403      0.4760
98  0.370     0.25890      0.4811
99  0.375     0.26460      0.4854
100 0.375     0.26382      0.4862

PIPS documentation built on May 1, 2019, 10:37 p.m.