Error missing value where TRUE/FALSE needed

[luis-zapata-b]

Whenever i try to run the code:

nlcor(x1,y1,plot=T) i get this error:

Error in if (sign(segment.cor$cor[i - 1]) == sign(segment.cor$cor[i])) { : Error missing value where TRUE/FALSE needed

When i put plot=F, it works but this warning mantains: Warning message: In FindSegmentSize(l = length(x), refine = refine) : Refinement too high or data is small. Adjusting computation.

Please help, i really need the plots.

[cran2367]

@luis-zapata-b Please try now. Run the following:

install.packages("devtools")
library(devtools)
install_github("ProcessMiner/nlcor")

library(nlcor)
library(ggplot2)

nonlinear.cor <- nlcor(x1, y1, plt = T)
print(nonlinear.cor$cor.plot)

Please let us know if the problem still exist.

[Jinstats]

I have a same problem as this.

[cran2367]

@Jinstats @luis-zapata-b Please replace plot = T with plt = T. The argument name here is plt.

[cran2367]

@Jinstats @luis-zapata-b please respond if the problem still exists. This issue will be otherwise closed in 5 days.

[luis-zapata-b]

I use plt, it must have been the autocorrector. I still have the same problem. Whenever it works with the x1 and y1 database, when i use my own data base (without NAs and both the same size), i keep getting the same error.

Please help

[Jinstats]

I exactly use the argument "plt=T", but still have the error.

[cran2367]

@luis-zapata-b @Jinstats did it work with the default x1 and y1 available with the package for testing? If it did and it does not work for your data, please share it with us so that we can reproduce the error.

[Jinstats]

default x1 and y1 works. Attached please find the data I used. test.xlsx

[cran2367]

@Jinstats Thank you for sharing the data. There was a bug that could not plot if the estimated relationship was linear. It is fixed and deployed. Please run it as:

install.packages("devtools")
library(devtools)
install_github("ProcessMiner/nlcor")

library(nlcor)
library(ggplot2)

data <- read.csv(file='test.csv', header = T)
nonlinear.cor <- nlcor(data$x1, data$y1, plt = T)
print(nonlinear.cor$cor.plot)

@luis-zapata-b Please try it. Hopefully this also solves your issue. If not, please let us know and share your data.

[cran2367]

@Jinstats @luis-zapata-b when you install the nlcor package please remove your current version with remove.packages('nlcor') and install again. Make sure the nlcor version you have reinstalled is nlcor_1.1.0

[Jinstats]

Thanks so much. it works.

[cran2367]

Great. Closing the issue.