LONG RANGE CORRELATION IN HUMAN WRITINGS

We study long range correlation in various human writings by mapping them into a simple 1d random walk model. This approach allows us to obtain better quality scaling data than the traditional power spectrum methods. Optimally written computer programs seem to have the highest scaling exponent, i.e., close to that of ideal 1/f noise. Literature, on the other hand, leads to smaller exponents. The Bible, however, has the strangest correlation among the Roman letter writings examined.