predict.outference: Make predictions from an '"outference"' object

In shuxiaoc/outference: Valid Inference in Linear Regression Corrected for Outlier Removal

Description

This function gives predictions as well as confidence intervals for the regression surfaces and the prediction intervals from an "outference" object. The syntax is the same as predict.lm.

Usage

## S3 method for class 'outference'
predict(object, newdata, interval = c("none",
  "confidence", "prediction"), level = 0.95, alpha.tilde = NULL, ...)

Arguments

object	an object of class "outference".
newdata,	an optional data frame in which to look for variables with which to predict. If omitted, the fitted values are used. WARNING: making predictions for many new data points with interval = "confidence" or "prediction" can be time-consuming, since for each data point, the function needs to compute the truncation set and solve the roots for a truncated survival function.
interval,	type of interval calculation. If set to "none", then only point predictions are made; if set to "confidence", then this function returns confidence intervals for the regression surface; if set to "prediction", then this function returns the prediction intervals.
level,	confidence level, default to 0.95.
alpha.tilde,	an extra parameter between 0 and 1-level, which is used when computing prediction intervals. If left NULL, then this function searches the alpha.tilde that gives shortest prediction intervals. See also 'details'.
...,	other arguments.

Details

If alpha.tilde = NULL, then this function iterates over alpha.tilde in seq(0, 1-level, length.out = 100)[c(5, 25, 50, 75, 95)] and returns the results with the shortest prediction intervals.

Value

This function gives a vector of predictions or a matrix of predictions and intervals with column names fit, lwr, upr if interval is set.

Author(s)

Shuxiao Chen <[e-mail]>

See Also

outference for model fitting;

summary.outference for summaries;

coef.outference for extracting coefficients;

confint.outference for confidence intervals of regression coefficients;

plot.outference for plotting the outlying measure;

Examples

## Brownlee’s Stack Loss Plant Data
data("stackloss")
## fit the model
## detect outlier using Cook's distance with cutoff = 4
fit <- outference(stack.loss ~ ., data = stackloss, method = "cook", cutoff = 4)
## predictions at the first two observations
predict(fit, newdata = stackloss[1:2, ], interval = "none")
predict(fit, newdata = stackloss[1:2, ], interval = "confidence")
predict(fit, newdata = stackloss[1:2, ], interval = "prediction")

shuxiaoc/outference documentation built on July 8, 2019, 8:30 p.m.