I would like to fit a Binomial GLM on a certain dataset. Using glm(...,family=binomial) everything works fine however I would like to do it with the caret train() function. Unfortunately I get an unexpected error which I cannot get rid of.
library("marginalmodelplots")
library("caret")
MissUSA <- MissAmerica08[,c(2,4,6,7,8,10)]
formula <- cbind(Top10, 9-Top10)~.
glmfit <- glm(formula=formula, data=MissUSA, family=binomial())
trainfit <-train(form=formula,data=MissUSA,trControl=trainControl(method = "none"), method="glm", family=binomial())
The error I get is:
"Error : nrow(x) == length(y) is not TRUE"
caret doesn't support grouped data for a binomial outcome. You can expand the data into a factor variable that is binary (Bernoulli) data. Also, if you do that, you do not need to use family=binomial() in the call to train.
Max
Related
I would like to check if my model (standardized) residuals are normally distributed.
model <- lm(ratiopermonth ~ Greenspace, data = mydata)
qqline(rstandard(model))
But I got an error message:
plot.new has not been called yet
What is wrong?
qqline is to be used after qqnorm.
r <- rstandard(model)
qqnorm(r)
qqline(r)
I need to compute marginal effects out of a Generalized Linear Model (family=Poisson) estimated via the svyglm function from the R package survey for a subsample.
First, I declared the survey desgin with:
myDesisgn = svydesign(id=data$id, strata=data$strata, weights=data$sw, data=data)
Second, I estimated my model as:
fit = svyglm(y~ x1 +x2, design=myDesisgn, data=data, subset= x3 == 1, family= poisson(link = "log"))
Finally, when I want to get the Average Marginal Effect for, let's say, x1 I run:
summary(margins(fit, variables = "x1", design=myDesisgn))
... but I get the following error message:
"Error in h(simpleError(msg, call)) :
error in evaluating the argument 'object' in selecting a method for function 'summary': 'x' and 'w' must have the same length"
Running the following does not work either:
summary(margins(fit, variables = "x1", design=myDesisgn, subset=x3==1))
Solution:
summary(margins(fit, variables = "x1", design=myDesisgn[myDesisgn$variables$x3 == 1]))
Subsetting complex surveys leads to problems in the error estimation. When interested in a parameter for a specific subsample, one should use the desired subsample to estimate the parameter of interest and the full sample for the estimation of its error.
For example, svyglm(y~x, data=data, subset = z == 1) does exactly this (beta_hat estimated using observations for which z=1 and se(beta_hat) using the full sample).
Subsetting a svy design is possible and it keeps the original design information about number of clusters, strata. The code shown above is the "manual" way of doing so. Alternative one can directly rely on the subset.survey.design {survey} function.
myDesign_subset <- subset(myDesign, data$x3 == 1)
The two methods are equivalent and produce correct z-stats.
Not sure if this more of a statistics question but the closest similar problem I could find is here, although I couldn't get it to work for my case.
I am trying to develop a pooled, penalized logistic regression model. I used mice to create a mids object and then fit a model to each dataset using caret repeated cross-validation with elastic net regression (glmnet) to tune parameters. The fitted object is not of class "mira" but I think I fixed that by changing the object class with the right list items. The major issue is that glmnet does not have an associated vcov method, which is required by pool().
I would like to use penalized regression based on the amount of variables and uncertainty over which ones are the best predictors. My data consists of 4x numeric variables and 9x categorical variables of varying levels and I anticipate including interactions.
Does anyone know how I might be able to create my own vcov method or otherwise address this issue? I am not sure if this is possible.
Example data and code are enclosed, noting that I am not able to share the actual data.
library(mice)
library(caret)
dat <- as.data.frame(list(time=c(4,3,1,1,2,2,3,5,2,4,5,1,4,3,1,1,2,2,3,5,2,4,5,1),
status=c(1,1,1,0,2,2,0,0,NA,1,2,0,1,1,1,NA,2,2,0,0,1,NA,2,0),
x=c(0,2,1,1,NA,NA,0,1,1,2,0,1,0,2,1,1,NA,NA,0,1,1,2,0,1),
sex=c("M","M","M","M","F","F","F","F","M","F","F","M","F","M","M","M","F","F","M","F","M","F","M","F")))
imp <- mice(dat,m=5, seed=192)
control = trainControl(method = "repeatedcv",
number = 10,
repeats=3,
verboseIter = FALSE)
mod <- list(analyses=vector("list", imp$m))
for(i in 1:imp$m){
mod$analyses[[i]] <- train(sex ~ .,
data = complete(imp, i),
method = "glmnet",
family="binomial",
trControl = control,
tuneLength = 10,
metric="Kappa")
}
obj <- as.mira(mod)
obj <- list(call=mod$analyses[[1]]$call, call1=imp$call, nmis=imp$nmis, analyses=mod$analyses)
oldClass(obj) <- "mira"
pool(obj)
Produces:
Error in pool(obj) : Object has no vcov() method.
I am seeing some weird behavior with the stepAIC function in the MASS package when dealing with multinomial logistic models. Here is some sample code:
library(nnet)
library(MASS)
example("birthwt")
race.model <- multinom(race ~ smoke, bwt)
race.model2 <- stepAIC(race.model, k = 2)
In this case race.model and race.model2 have identical terms; stepAIC did not prune anything. However, I need to query certain attributes of the models, and I get an error with race.model2:
formula(race.model)[2]
returns race() but
formula(race.model2)[2]
gives the error:
Error in terms.formula(newformula, specials = names(attr(termobj, "specials"))) :
invalid model formula in ExtractVars
This behavior only seems to occur when stepAIC does not remove terms from the model. In the following code, terms are removed by stepAIC, and both models can be properly queried:
race.big <- multinom(race ~ ., bwt)
race.big2 <- stepAIC(race.big, k = 2)
formula(race.big)[2]
formula(race.big2)[2]
Any ideas about what is going wrong here?
I found that the predict function is currently not implemented in cumulative link mixed models fitted using the clmm function in ordinal R package. While predict is implemented for clmm2 in the same package, I chose to apply clmm instead because the later allows for more than one random effects. Further, I also fitted several clmm models and performed model averaging using model.avg function in MuMIn package. Ideally, I want to predict probabilities using the average model. However, while MuMIn supports clmm models, predict will also not work with the average model.
Is there a way to hack the predict function so that the function not only could predict probabilities from a clmm model, but also predict using model averaged coefficients from clmm (i.e. object of class "averaging")? For example:
require(ordinal)
require(MuMIn)
mm1 <- clmm(SURENESS ~ PROD + (1|RESP) + (1|RESP:PROD), data = soup,
link = "probit", threshold = "equidistant")
## test random effect:
mm2 <- clmm(SURENESS ~ PROD + (1|RESP) + (1|RESP:PROD), data = soup,
link = "logistic", threshold = "equidistant")
#create a model selection object
mm.sel<-model.sel(mm1,mm2)
##perform a model average
mm.avg<-model.avg(mm.sel)
#create new data and predict
new.data<-soup
##predict with indivindual model
predict(mm1, new.data)
I got the following error message:
In UseMethod("predict") :
no applicable method for predict applied to an object of class "clmm"
##predict with model average
predict(mm.avg, new.data)
Another error is returned:
Error in predict.averaging(mm.avg, new.data) :
predict for models 'mm1' and 'mm2' caused errors
I've been using clmm as well and yes I confirm predict.clmm is NOT (yet?) implemented. I didn't yet check the source code for fake.predict.clmm. It might work. If it doesn't, you're stuck with doing stuff by hand or using predict.clmm2.
I found a potential solution (pasted below) but have not been able to make work for my data.
Solution here: https://gist.github.com/mainambui/c803aaf857e54a5c9089ea05f91473bc
I think the problem is the number of coefficients I am using but am not experienced enough to figure it out. Hopefully this helps someone out though.
This is the model and newdata that I am using, though it is actually a model averaged version. Same predictors though.
ma10 <- clmm(Location3 ~ Sex * Grass3 + Sex * Forb3 + (1|Tag_ID), data =
IP_all_dunes)
ma_1 <- model.avg(ma10, ma8, ma5)##top 3 models
new_ma<- data.frame(Sex = c("m","f","m","f","m","f","m","f"),
Grass3 = c("1","1","1","1","0","0","0","0"),
Forb3 = c("0","0","1","1","0","0","1","1"))
# Arguments:
# - model = a clmm model
# - modelAvg = a clmm model average (object of class averaging)
# - newdata = a dataframe of new data to apply the model to
# Returns a dataframe of predicted probabilities for each row and response level
fake.predict.clmm <- function(modelAvg, newdata) {
# Actual prediction function
pred <- function(eta, theta, cat = 1:(length(theta) + 1), inv.link = plogis) {
Theta <- c(-1000, theta, 1000)
sapply(cat, function(j) inv.link(Theta[j + 1] - eta) - inv.link(Theta[j] -
eta))
}
# Multiply each row by the coefficients
#coefs <- c(model$beta, unlist(model$ST))##turn off if a model average is used
beta <- modelAvg$coefficients[2,3:12]
coefs <- c(beta, unlist(modelAvg$ST))
xbetas <- sweep(newdata, MARGIN=2, coefs, `*`)
# Make predictions
Theta<-modelAvg$coefficients[2,1:2]
#pred.mat <- data.frame(pred(eta=rowSums(xbetas), theta=model$Theta))
pred.mat <- data.frame(pred(eta=rowSums(xbetas), theta=Theta))
#colnames(pred.mat) <- levels(model$model[,1])
a<-attr(modelAvg, "modelList")
colnames(pred.mat) <- levels(a[[1]]$model[,1])
pred.mat
}