Bernoulli Numbers are defined as those B_{k} in the following equation: [1.1] We
need to expand the left-hand side and equate the coefficients of the
power series with corresponding coefficients of the right-hand side.

At
first sight, it seems difficult to expand the left-hand side into a
power series using Maclaurin's expansion. In fact there are many ways
of expanding the left-hand side and we will mention two of them.
Actually, we need to derive a formula which makes the calculation
easier, or we can use the symbolic/mnemonic method.

In either case we need: [1.2]

And, expanding the RHS: [1.3]

The
first method we could use is to divide x in 1.1 by 1.2, using
normal polynomial division. What we need to remember is that we need to
have sufficient terms of e^{x}-1 to get our required coefficients.

The second method being mentioned is simply to multiply the RHS of 1.1 by (e^{x}-1), and equate the coefficients of the x's. [1.4]

Expanding both sides: [1.5]

Now we can equate coefficients:

Equating

[1.5.1]

x

[1.5.2]

x^{2}

[1.5.3]

x^{3}

[1.5.4]

x^{4}

Because
we kept all our factorials, a patterns becomes evident. Except
for 1.5.1 we have the formula, which can be used to generate the
Bernoulli Numbers. In the formula below, n is the highest Boolean
number in the series (power of x is n+1). So for a series ending in B_{n}, we have: [1.6]

For
convenience, we can multiply throughout by (n+1)! (because we see how the
bottom factorials seem to be related to a combination, ^{n+1}C_{k}) in order to get a formula for generating the Bernoulli Numbers: [1.7]

Ken's book is packed with examples and explanations that enable you to discover more than 150 techniques to speed up your arithmetic and increase your understanding of numbers. Paperback and Kindle: