I am confused by this warning message as I try to fit my data with a nonlinear regression model by using the drc package and drm function.
I have
N_obs <- c(1, 80, 80, 80, 81, 82, 83, 84, 84, 95, 102, 102, 102, 103, 104, 105, 105, 109, 111, 117, 120, 123, 123, 124, 126, 127, 128, 128, 129, 130)
times <- c(3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32)
The model is
model.drm <- drm(N_obs ~ times, data = data.frame(N_obs = N_obs, times = times), fct = MM.2())
and the warnings come from predictions
preds <- predict(model.drm, times = times, interval = "confidence", level = 0.95)
There were 30 or more warnings (use warnings() to see the first 50)
> warnings()
Warning messages:
1: In (tquan * sqrt(varVal + sumObjRV)) * c(-1, 1) :
Recycling array of length 1 in array-vector arithmetic is deprecated.
Use c() or as.vector() instead.
2: In (tquan * sqrt(varVal + sumObjRV)) * c(-1, 1) :
Recycling array of length 1 in array-vector arithmetic is deprecated.
Use c() or as.vector() instead.
3: In (tquan * sqrt(varVal + sumObjRV)) * c(-1, 1) :
Recycling array of length 1 in array-vector arithmetic is deprecated.
Use c() or as.vector() instead.
I have been trying to change data inputs by using as.vector(times), c(times),etc., but still cannot get rid of the warnings. Could someone help me identify the problem? Thank you!!
I re-ran your analysis with the sample data provided, and I can reproduce your warnings. Here's a summary:
Fit a Michaelis-Menten model of the form f(x; d, e) = d * (1 + e/x)^-1
.
# Fit a 2 parameter Michaelis-Menten model library(drc); fit <- drm( formula = N_obs ~ times, data = data.frame(N_obs = N_obs, times = times), fct = MM.2())
Based on the model fit, predict
the response for the original times
. Note you can omit the newdata
argument here, because in that case predict
will simply use the fitted values (which are based on times
).
# Predictions pred <- as.data.frame(predict( fit, newdata = data.frame(N_obs = N_obs, times = times), interval = "confidence", level = 0.95)); pred$times <- times;
Visualise data and predictions.
library(tidyverse); data.frame(times = times, N_obs = N_obs) %>% ggplot(aes(times, N_obs)) + geom_point() + geom_line(data = pred, aes(x = times, y = Prediction)) + geom_ribbon( data = pred, aes(x = times, ymin = Lower, ymax = Upper), alpha = 0.4);
The model fit seems reasonable, and I would say that the warnings
can be safely ignored (see details).
I had a look at the drc
source-code, and the warning originates from line 201
of predict.drc.R
:
retMat[rowIndex, 3:4] <- retMat[rowIndex, 1] + (tquan * sqrt(varVal + sumObjRV)) * c(-1, 1)
In that line, an array of dimension 1 is added to a numeric vector.
Here is a simple example to reproduce the warning:
arr <- array(5, dim = 1);
arr + c(1, 2);
#[1] 6 7
#Warning message:
#In arr + c(1, 2) :
# Recycling array of length 1 in array-vector arithmetic is deprecated.
# Use c() or as.vector() instead.
Note that the result is still correct; it's just that R doesn't like the addition of a one-dimensional array and a vector, and prefers instead adding proper scalars and vectors, or vectors and vectors.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.