For individual courses and professors you need to ask the professor. However, there are a number of reasons why someone might want to make no recommendation.
Least likely is that the prof feels that any book is as good as any other.
A bit more likely is that the prof feels that any book is as bad as any other.
Up the scale a bit is the sense (possibly misplaced) that the prof's lectures are all the student will need. If this is the case, the student's are strongly advised to attend every lecture and take lots of notes. For some this is a valid position if the professor also puts extensive materials online or otherwise makes them available. However, it can also be a trap if the professor thinks that lectures deliver the needed information and skills perfectly to every student. That is a serious error of judgement.
Another reason is that the prof wants the students to actually seek out answers to questions online or in the (gasp) library. Some professors don't answer questions with answers but with a strategy for finding the answer. This, of course, disadvantages lazy students.
Still higher on the (my) list is that the course is intended to use active learning and so more passive approaches (reading, watching, listening) are discouraged and the student is expected to do most of his or her learning by doing exercises and projects. For Statistics and Numerical Analysis, this seems to me to be a worthy goal. I find it useful also in much of computer science.
Given that one learns by practice and reinforcement, this last strategy can be very effective. Active learning gives you an operational knowledge of a subject that reading (or even reading and underlining) a book won't.
I have, on occasion, "recommended" a book, not for help on the things in the course, but for things that won't be covered. There was no obligation to buy the book. I've also made such recommendations about a pair of quite different books for the same reason.