'why a specific model is not appropriate, given a data with 6 variables (they are chr variables)
i want to show why a specific model is not appropriate, given a data with 6 variables (they are chr variables) the model is y= abc*(x1+x2) a and b from the data are non numeric, so is it enough to say that because they are not numeric we cant use this model? and if i want to change a chr variables into numeric how can i do this without making it become NA?
Solution 1:[1]
You can still potentially model non-numeric type data by converting it to factor by: df$a <- as.factor(df$a); df$b <- as.factor(df$b).
However, it might not make sense sometimes. For example, df$a is an unique id for each observation and you don't want to put it as a predictor.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Fujibayashi Kyou |