Getting an error 'Error in -2 * var_cll : non-numeric argument to binary operator'

[ArjunReddy07]

Hi,

I am running 'vinereg' for different sets of data. I have one response variable and three independent variables ( I do not explicitly know the relation between them).

• The code used looks like the below, each variable is a column of size 911. av <-data.frame(y=input_data_raw, x1=input_data_raw_1, x2=input_data_raw_2, x3=input_data_raw_3) vine_reg_model_raw_data <- vinereg(y ~ ., av)

• The nature of the data is such that y takes the column of numbers between (-1,1) and the independent variables take the values between (0, 1).

• For some sets of the initialized y, x1, x2, and x3 (each of them are columns of size 911), there is no error in the QR, but for some sets of initialized data, I am getting the following error '*Error in -2 var_cll : non-numeric argument to binary operator**'

• A traceback is showing the following '2: finalize_vinereg_object(formula = formula, selcrit = selcrit, model_frame = mf, margins = margins, vine = fit$vine, selected_vars = fit$selected_vars, var_nms = colnames(mfx)) 1: vinereg(y ~ ., av)'

I am not able to understand the cause for such an error, would be helpful if resolved. Thank you.

[tnagler]

Is it possible to give a reproducible example? (You can also send me an email if you don't want to post data publicly.) I have an idea what is happening, but without example, it's very hard to find out why.

[ArjunReddy07]

Ok sure, thank you.

[ArjunReddy07]

Please use the attached code snippet and the data from the excel file. This is one instance where there is an error. For some other similar instances of data, 'vinereg' is working fine. Please let me know if any concerns.

Sincerely, Arjun

On Wed, 24 Feb 2021 at 13:31, Thomas Nagler notifications@github.com wrote:

Is it possible to give a reproducible example? (You can also send me an email if you don't want to post data publicly.) I have an idea what is happening, but without example, it's very hard to find out why.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/tnagler/vinereg/issues/9#issuecomment-784877043, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMBTNMMSH2QIALMYETUR7CTTASW65ANCNFSM4YD4RKRA .

[tnagler]

There's no attachment, please send to mail@tnagler.com.

[tnagler]

Alright, so in your example, no covariates were selected to have predictive power. This case wasn't handled properly in the library, should be fixed now. You can install an updated version with

devtools::install_github("tnagler/vinereg")

I'll submit the new version to CRAN soon.

[tnagler]

Alright, so in your example, no covariates were selected to have predictive power. This case wasn't handled properly in the library, should be fixed now. You can install an updated version with

devtools::install_github("tnagler/vinereg")

I'll submit the new version to CRAN soon.

[ArjunReddy07]

ok, I was not able to think about this case. Thank you for the help. I will install the updated version and try again.

On Wed, 24 Feb 2021 at 19:13, Thomas Nagler notifications@github.com wrote:

Alright, so in your example, no covariates were selected to have predictive power. This case wasn't handled properly in the library, should be fixed now. You can install an updated version with

devtools::install_github("tnagler/vinereg")

I'll submit the new version to CRAN soon.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/tnagler/vinereg/issues/9#issuecomment-785084325, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMBTNMJ6H6CAR7WV7WBTOR3TAT7BJANCNFSM4YD4RKRA .