Closed omzeybek closed 2 years ago
Hi, thank you for your interest in the package.
I run the auto_ardl()
function in debug mode and I managed to reproduce the following error:
Error in list(order2_back, order2_forth)[which(c(selection_m2_back, selection_m2_forth) == : subscript out of bounds
Assuming that this is the cause of your problem too, here is what I think.
-Inf
. This makes the auto_ardl()
to select this model as this will always have the smallest criterion (but this will not cause a problem like in your case).NA
. Not sure if and why this NA
is produced by the AIC/BIC or whatever the criterion is.Not sure if this is indeed the problem. If it is, it is primarily a problem with the criterion. Naturally, an NA
value can not be compared with another numeric value, and so the auto_ardl()
function can not decide which one to choose.
I hope you find a workaround for this problem. In any case, lets me know if you have any new insights.
As for the warnings, I'm not sure that is the problem here. It seems like it can't handle some factor variables, but again this has to do with the regression model and not with the package. A small dataset would be helpful so I can reproduce it.
Regards
Hi,
I am very interested in your ARDL package. To understand it's components I ran data(denmark) example you gave in documentation. Everything was fine with example data set, except exogenous variable issue I asked in my previous question. But when I turned to my real data which in zooreg data frame type and containing numeric variables. I started receving the following error. Both tow variables I used for analysis doesn't includes missing values and they are in logarithmic form.
models <- auto_ardl(gunluk_miktar~ promotion_depth,data=dfzooreg,max_order = 2)
I am not sure what's the problem but I am receving the error message bellow.
Error in list(order2_back, order2_forth)[which(c(selection_m2_back, selection_m2_forth) == : subscript out of bounds In addition: Warning messages: 1: In model.response(mf, "numeric") : using type = "numeric" with a factor response will be ignored 2: In Ops.factor(y, z$residuals) : ‘-’ not meaningful for factors 3: In Ops.factor(res, 2) : ‘^’ not meaningful for factors 4: In model.response(mf, "numeric") : using type = "numeric" with a factor response will be ignored 5: In Ops.factor(y, z$residuals) : ‘-’ not meaningful for factors 6: In Ops.factor(res, 2) : ‘^’ not meaningful for factors 7: In model.response(mf, "numeric") : using type = "numeric" with a factor response will be ignored 8: In Ops.factor(y, z$residuals) : ‘-’ not meaningful for factors 9: In Ops.factor(res, 2) : ‘^’ not meaningful for factors
If it's possible I kindly need a help on error handling
thank you very much for your help