Roberts, A.Cronin, A.2011-08-162011-08-161996Physica A: Statistical Mechanics and its Applications, 1996; 233(3-4):867-8780378-4371http://hdl.handle.net/2440/65431We present a novel method for determining multi-fractal properties from experimental data. It is based on maximizing the likelihood that the given finite data set comes from a particular set of parameters in a multi-parameter family of well known multi-fractals. By comparing characteristic correlations obtained from the original data with those that occur in artificially generated multi-fractals with the same number of data points, we expect that predicted multi-fractal properties are unbiased by the finiteness of the experimental data.enCopyright © 1996 Published by Elsevier B.V.Finite data setsmulti-fractal spectrumbinary multiplicative multi-fractalgeneralized dimensionsUnbiased estimation of multi-fractal dimensions of finite data setsJournal article00300010202011081613550310.1016/S0378-4371(96)00165-3A1996VZ257000252-s2.0-003040096165026Roberts, A. [0000-0001-8930-1552]