Short answer: I don’t know.
First, to be honest, I don’t know whether I can apply my usual understanding of the word ’caused’ to this problem at all. I think of cause as something that connects events A and B when A occurred before B and B’s occurrence was dependent on A happening.
The trouble is, before the Universe, it’s far from clear that there’s any clear way to talk about before. I can only think about time as a property of spacetime, which came into existence at the moment of the Big Bang, as far as we know. Without spacetime, I don’t believe the concepts of distance or duration (necessary to think about time) can exist.
It’s entirely possible that my understanding or Time or the Big Bang is flawed, but, without an alternate, plausible conception of either, I can’t make much headway on this question.
The Big Bang itself seems fairly plausible to me (with my limited understanding of the physics involved. I owe much of my understanding to Simon Singh’s Big Bang. It’s a fascinating read, and the kind of evidence it presents (even dumbed down for laypeople) helps to make it clear how science really works. The Big Bang theory didn’t match the prevailing observations about the world, it exceeded them. It didn’t surpass them by offering a more compelling mechanism causing the cosmological events that had been observed to date (since a sufficiently creative person could make up an interesting story to explain almost any set of facts), but by making predictions about observations that had never been made, predictions that conflicted with the predictions of the Steady State model.
I mention all this, because this standard of evidence is similar to the one I subject Christianity to, and I’ve found it wanting. The primary proof I’ve been offered for Christianity is that it offers a fuller explanation for morality than I’ve come been able to come up with myself. However, as long as Christianity only explains pre-observed phenomena, it’s explanatory power is not persuasive evidence alone. Either there must be evidence for the mechanism by which it works, or it should be able to make predictions whose truth values are not yet known, but turn out to be true upon testing.
Otherwise, I see no reason to give up truthful uncertainty for the sake of having an explanation, if the explanation doesn’t go beyond what I already know.