I'm trying to predict if a company will bankrupt, I use a dataset of 2020 and I manually created my target variable with the status of the company the status date, status reason to create my target variable. Could I use these variables too for my model or because I build my target with it it's totally forbidden ? (My opinion is that I cannot but I'm curious to hear what the community will say) Thanks.
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
If you would already have those datapoints before a company actually goes into bankruptcy then you can then them in your model since when predicting to the future you could have access to that data. However, if you would only know the data once the bankruptcy event happens (e.g. date of bankruptcy) then you cannot use this variable in your model since you would be leaking data (using data that is in the future and the model would not have access to when the model is actually deployed and used).
