Power of a negative decimal raised to a decimal

Question

the Javadoc of java.lang.Math.pow(double a, double b) states:

• If the first argument is finite and less than zero:
  • if the second argument is finite and not an integer, then the result is NaN.

This means a call to Math.pow(-Math.E,-1.1d) yields NaN.

Why isn't Math.pow() returning the inverse 1/e^1.1? Is there an error in my reasoning?

Thanks!

Answer 1

Yes, why it should return 1/e^1.1? This will be 1/(-e)^1.1 which is not real number.

Answer 2

Yes, there's an issue with your logic. Please go and read about complex numbers.

The problem is that a negative base raised to a non-integer negative power results in a complex number, not a real double. There's an imaginary part that Math.pow can't deal with.