When dealing with an empiricist who wants evidence that God exists, and yet thinks evidence—for it to be considered evidence—must be empirical in nature, ask him the following question: “What kind of empirical evidence could possibly be given for an immaterial being such as God?”  If they say “none,” then point out that they are asking for the impossible.  What would it prove, then, if you cannot deliver?  Nothing.  It just proves that the wrong question is being asked.

Insisting on empirical evidence before one will believe in the existence of God is like insisting on chemical evidence of your wife’s love for you before you’ll believe she loves you.  One cannot supply chemical proof for love, and neither can one supply empirical proof of God’s existence, but that does not mean either is false.  The problem is not a lack of evidence for God’s existence, but an arbitrary restraint on the kind of evidence the atheist is willing to accept as evidence.  That is what needs to be challenged.  Empirical evidence is not the only kind of evidence one can appeal to in support of a claim.

I should point out, however, that while there cannot be direct empirical evidence for an immaterial being, we can infer the existence of such a being from the effects it may cause in the material world.  For example, we can infer the existence of God from the empirical evidence that the universe began to exist since anything which begins to exist requires an external cause, and God is the only kind of causal entity external to the universe that is sufficient to explain the effect in question.

See also Empirical Evidence and the Existence of God.