Hi - I hope someone can share some insight on how to interpret the following regression analysis result.

I have been analysing the demand for advertising on radio stations in relation to their audience share in a specific market. The result is a very significant linear regression, however the formula predicts that the demand turns negative when audience share is below 2%. BTW, there were no samples of such radio stations (<2%) included as inputs for the regression.

Anecdotally, the media buyers in the market have always said they have no interest in buying advertising on small stations (approximately <2% share). But can the regression result really be interpreted as a support for this hypothesis?

Any help/opinion would be appreciated.