Round 3 Data

Dr. Ballings,

One of the 300 stocks you are asking us to predict for round 3 is called 'BCH'. We have no data on this stock to this point.

How should we go about handling this? Should we just assign it a 0 probability since it doesn't exist up to this point?

Or should we look at our other predictions for this round, and get a prediction for 'BCH' by imputing it?

Or should we assume it's an error and go with one of the other stocks beginning with 'BC'?


Answers and follow-up questions

Answer or follow-up question 1

Dear Michael,

That is well possible. Companies are created and go out of business every day.

How to cope with this?

Set the prediction for that stock equal to mean(yourpredictions).
In this case yourpredictions are the remainder of the 300 stocks.

Michel Ballings

Sign in to be able to add an answer or mark this question as resolved.