Take the following case: In discussing why determiner phrases (DPs) include NPs as arguments and not the other way round, Progovac puts forward as a basic assumption that "there is systematic parallelism between sentences and NPs (see (2)), which can best be captured by introducing functional categories into NPs" (p. 166), giving the following example:
(2) (a) [ S John builds spaceships]The problem is that 2b has a clause as its subject, not an NP, despite the notation (the S in 2a is for "sentence", but "clause" would be better.) It's not a typical clause by any means, but it clearly has a subject-predicate structure, despite its subject being in the genitive case rather than the nominative typical of most subjects, and despite the non-finite nature of the clause. (The arguments are set out in CGEL, pp. 1189–1190.) So really what Progovac says here is that there is a systematic parallelism between clauses and clauses. And, of course, this bears not at all on the structure of NPs or DPs for that matter.
(b) [ NP John’s building a spaceship] upset me.
This doesn't mean it's a bad idea, just that it's a bad example. For more on the idea, see this post.
Huddleston, R., & Pullum, G. K. (2002). The Cambridge grammar of the English language. Cambridge: Cambridge Univ Press.
Progovac, L. (1998). Determiner phrase in a language without determiners (with apologies to Jim Huang 1982). Journal of Linguistics, 34(1), 165–179.