Predicting from Nonlinear Least Squares Fits
predict.nls produces predicted values, obtained by evaluating
the regression function in the frame
newdata. If the logical
TRUE, standard errors of the predictions are
calculated. If the numeric argument
scale is set (with
df), it is used as the residual standard deviation in
the computation of the standard errors, otherwise this is extracted
from the model fit. Setting
intervals specifies computation of
confidence or prediction (tolerance) intervals at the specified
interval are ignored.
## S3 method for class 'nls': predict(object, newdata , se.fit = FALSE, scale = NULL, df = Inf, interval = c("none", "confidence", "prediction"), level = 0.95, ...)
- An object that inherits from class
- A named list or data frame in which to look for variables with
which to predict. If
newdatais missing the fitted values at the original data points are returned.
- A logical value indicating if the standard errors of the
predictions should be calculated. Defaults to
FALSE. At present this argument is ignored.
- A numeric scalar. If it is set (with optional
df), it is used as the residual standard deviation in the computation of the standard errors, otherwise this information is extracted from the model fit. At present this argument is ignored.
- A positive numeric scalar giving the number of degrees of
freedom for the
scaleestimate. At present this argument is ignored.
- A character string indicating if prediction intervals or a confidence interval on the mean responses are to be calculated. At present this argument is ignored.
- A numeric scalar between 0 and 1 giving the confidence level for the intervals (if any) to be calculated. At present this argument is ignored.
- Additional optional arguments. At present no optional arguments are used.
predict.nlsproduces a vector of predictions. When implemented,
intervalwill produce a matrix of predictions and bounds with column names
upr. When implemented, if
TRUE, a list with the following components will be returned:
fit vector or matrix as above se.fit standard error of predictions residual.scale residual standard deviations df degrees of freedom for residual
Variables are first looked for in
newdata and then searched for
in the usual way (which will include the environment of the formula
used in the fit). A warning will be given if the
variables found are not of the same length as those in
if it was supplied.
<testonly>od <- options(digits = 5)</testonly> require(graphics) fm <- nls(demand ~ SSasympOrig(Time, A, lrc), data = BOD) predict(fm) # fitted values at observed times ## Form data plot and smooth line for the predictions opar <- par(las = 1) plot(demand ~ Time, data = BOD, col = 4, main = "BOD data and fitted first-order curve", xlim = c(0,7), ylim = c(0, 20) ) tt <- seq(0, 8, length = 101) lines(tt, predict(fm, list(Time = tt))) par(opar) <testonly>options(od)</testonly>