3 ms·
Just because the error interval is [-inf, +inf] doesn't mean the actual error is [-inf, +inf]. There are other techniques (such as Monte Carlo simulation) that
by jcranmer 8y ago
Just because the error interval is [-inf, +inf] doesn't mean the actual error is [-inf, +inf]. There are other techniques (such as Monte Carlo simulation) that can give much narrower error bounds for the actual computation.