Page 105 - Standard Handbook Of Petroleum & Natural Gas Engineering
P. 105

94    Mathematics
                                               Moment Ratios

                      The moment  ratios  are dimensionless coefficients used to compare characteristics
                    of  distributions  measured  on  different  scales.
                      The coefficient  of  variation is a measure of relative dispersion  of  a set of  values
                    and is  given  for  the  population  by



                    and for  the  sample by

                      C" = s/x

                      The coefficient  of  skewness  is a measure of  relative symmetry of  a distribution
                    and  is  given  for  the  population  by




                    and for  the  sample by




                      The coefficient  of  kurtosis  is a measure  of  relative peakedness  and  is  given by
                      P,  = k/S4

                       Common  Probability Distributions for  Continuous Random Variables
                      The parameters  of  a distribution  control  its  geometric  characteristics  [23]:

                      1. A location parameter is the abscissa of  a location point and may  be a measure
                         of  central  tendency,  such  as  a mean.
                      2. A  scale  parameter  determines  the  location  of  fractiles  of  the  distribution
                         relative to some specified  point, often the value of  the location parameter.
                      3.  Shape  parameters  control  the  geometric  configuration  of  a  distribution.
                         There  may  be  zero,  one,  or multiple  shape parameters.

                      A  bounded  continuous  random  variable  with  uniform  distribution  has  the
                    probability  function
                       f(X)  = I $'(P-a)  a < X < P
                                       otherwise

                    where  a = location parameter,  representing  lower limit  of  the  distribution
                          P  = scale parameter,  representing  upper  bound  of  the  distribution

                    Probabilities  are determined  by  integration  over  the  necessary  range,  Le.,
                       P(X, < x < X,) = p/(p - a)dX
                                       I
   100   101   102   103   104   105   106   107   108   109   110