I ran into the following error when running the sem() function:
I have just ran this code using a different syntax and it worked; so I know the problem should be either in the way I loaded the factors to the syntax... or a typo?
library(psych)
library(sem)
#Creating syntax equation
mgb_syn_eq <- "
MIL: S.Defense #Military and National Security
GOV: E.GovLess, E.GovFM, E.GovBig #Limited Government
BUS: E.BusinessReg #Business
"
#Creating CFA Syntax from theory
mgb_syn <- cfa(text = mgb_syn_eq, reference.indicators = FALSE)
#Use the sem() function to run a CFA
mgb_CFA <- sem(mgb_syn, data = df.secs_CFA)
Error in w_mat %% p_deriv_mat %% invMat : requires numeric/complex matrix/vector arguments
I found the answer:
The problem was that psych package only has some of the basic functions of the lavaan package AND that I was trying to estimate a latent value with only 1 parameter (you need at least 3).
Related
I am trying to implement ordinal logistic regression on my dataset in r. I use the function 'polr' for this, but cannot seem to find a lot of information regarding its implementation.
The following errors are the ones I'm stuck on:
> dat.polr <- polr(as.factor(relevance)~allterms+idf.title, data=dat.one)
Warning message:
In polr(as.factor(relevance) ~ allterms + idf.title + idf.desc + :
design appears to be rank-deficient, so dropping some coefs
> dat.pred <- predict(dat.polr,dat.test,type="class")
Error in X %*% object$coefficients : non-conformable arguments
I want to train my model to guess the relevance of a new dataset. dat.one is the dataset I'm using to train the data, dat.test is the dataset I'm using to test the data. I believe that the predict variable's error is caused by the warning in polr. However, I have no clue how to resolve this. Any help would be appreciated :)
I am estimating a fairly simple McFadden choice model using a very large data set (101.6 million unit-alternatives). I can estimate this model just fine in Stata using the asclogit command, but when I try to use the mlogit package in R, I get the following error:
region1 <- mlogit(chosen ~ mean_log.wage + mean_log.rent + bornNear + Dim.1 + regionFE | 0,
shape= "long", chid.var = "chid", alt.var = "alternatives", data = ready)
Error in qr.default(na.omit(X)) : too large a matrix for LINPACK
Calls: mlogit ... model.matrix -> model.matrix.mFormula -> qr -> qr.default
If I look at the source code of qr.R it's clear that the number of elements in my design matrix is too big relative to the LINPACK limit of 2,147,483,647. However, no such limit exists for LAPACK (that I can tell, at least).
From qr.R:
qr.default <- function(x, tol = 1e-07, LAPACK = FALSE, ...)
{
x <- as.matrix(x)
if(is.complex(x))
return(structure(.Internal(La_qr_cmplx(x)), class = "qr"))
## otherwise :
if(LAPACK)
return(structure(.Internal(La_qr(x)), useLAPACK = TRUE, class = "qr"))
## else "Linpack" case:
p <- as.integer(ncol(x))
if(is.na(p)) stop("invalid ncol(x)")
n <- as.integer(nrow(x))
if(is.na(n)) stop("invalid nrow(x)")
if(1.0 * n * p > 2147483647) stop("too large a matrix for LINPACK")
...
qr() appears to be called in the mFormula method of mlogit, when model.matrix is being created, and probably while checking NAs. But I can't tell if there is a way to pass LAPACK = TRUE to mlogit, or if there is a way to skip the NA checking.
I'm hoping #YvesCroissant will see this.
As I mentioned, I can estimate this model just fine in Stata, so it's not a question of resources. My Stata license is not portable, however, which is why I would like to use R.
Thanks to Julius' comment and this post on namespaces in R, I figured out the answer. I added the following code right after my library statements:
source("mymFormula.R")
tmpfun <- get("model.matrix.mFormula", envir = asNamespace("mlogit"))
environment(mymFormula) <- environment(tmpfun)
attributes(mymFormula) <- attributes(tmpfun) # don't know if this is really needed
assignInNamespace("model.matrix.mFormula", mymFormula, ns="mlogit")
mymFormula.R is an R script where I copy/pasted the contents of mlogit:::model.matrix.mFormula and added mymFormula <- before the function invocation at the top of the file.
I viewed the contents of mlogit:::model.matrix.mFormula by typing trace(mlogit:::model.matrix.mFormula, edit=TRUE) in RStudio. (Thanks to this answer for help on how to do that.)
In package rugarch,
I want to list out the AIC of several garch models in R.
But I don't know how to to :(, So I use the loop for to get the infocriteria of them. my code below:
for(i in 1:5)
{
for(j in 1:5)
{
garch11.spec=ugarchspec(variance.model=list(garchOrder=c(i,j)))
garch11.fit=ugarchfit(spec=garch11.spec, data=google_rets)
print(infocriteria(garch11.fit))
}
}
After running 2 models, an error occupts:
Error in itestm[1, 1] <- itest$AIC : replacement has length zero
In addition: Warning messages:
1: In .sgarchfit(spec = spec, data = data, out.sample = out.sample, :
ugarchfit-->warning: solver failer to converge.
2: In log(log(nObs)) : NaNs produced
How can I solve the error?
Could you give me a way to get the AIC from package rugrach
You may use auto.arima() and arimaorder() to get the p,d,q values and fit into the ugarchspec(), it will be more effcient than using for loop...
You can use google translate and refer to binary.com 面试试题 I - GARCH模型中的ARIMA(p,d,q)参数最优化 for more information.
I'm trying to use the gamsel R package to fit a sparse generalized additive model, and I can't seem to get it to work on real data. When I run on synthetic data as described in the package documentation, everything works well:
library(gamsel)
data=gendata(n=500,p=12,k.lin=3,k.nonlin=3,deg=8,sigma=0.5)
attach(data)
bases=pseudo.bases(X,degree=10,df=6)
gamsel.out=gamsel(X,y,bases=bases)
But when I run on real data, I get the following error:
library(gamsel)
X = as.matrix(read.csv("X.csv"),header=FALSE)
y = as.matrix(read.csv("y.csv"),header=FALSE)
gam_fit = gamsel(X,y)
Error in if (abs((df - current.df)/df) < 1e-04 | iterations == 1)
return(list(lambda = lambda, : missing value where TRUE/FALSE
needed
You can access sample data files that will reproduce this result here. Any thoughts about how to fix this error?
I am getting an error while running naive bayes classifier in R. I am using the following code-
mod1 <- naiveBayes(factor(X20) ~ factor(X1) + factor(X2) +factor(X3) +factor(X4)+factor(X5)+factor(X6)+factor(X7)
+factor(X8)+factor(X9)
+factor(X10)+factor(X11)+ factor(X12)+factor(X13)+factor(X14)
+factor(X15)
+factor(X16)+factor(X17)
+factor(X18)+factor(X19),data=intent.test)
res1 <- predict(mod1)$posterior
First part of this code runs fine. But when it try to predict the posterior probability it throws following error-
**Error in as.data.frame(newdata) :
argument "newdata" is missing, with no default**
I tried running something like
res1 <- predict(mod1,new_data=intent.test)$posterior
but this also gives the same error.
You seem to be using the e1071::naiveBayes algorithm, which expects a newdata argument for prediction, hence the two errors raised when running your code. (You can check the source code of the predict.naiveBayes function on CRAN; the second line in the code is expecting a newdata, as newdata <- as.data.frame(newdata).) Also as pointed out by #Vincent, you're better off converting your variables to factor before calling the NB algorithm, although this has certainly nothing to do with the above errors.
Using NaiveBayes from the klar package, no such problem would happen. E.g.,
data(spam, package="ElemStatLearn")
library(klaR)
# set up a training sample
train.ind <- sample(1:nrow(spam), ceiling(nrow(spam)*2/3), replace=FALSE)
# apply NB classifier
nb.res <- NaiveBayes(spam ~ ., data=spam[train.ind,])
# predict on holdout units
nb.pred <- predict(nb.res, spam[-train.ind,])
# but this also works on the training sample, i.e. without using a `newdata`
head(predict(nb.res))