Titanic Fares

Here we have the fares of passengers of the Titanic.

>>> from effectus import Effects
>>> from effectus.data import titanic_fares
>>> tf = Effects(titanic_fares())
>>> tf
<pareto present [0.96]: 1/5 causes => 2/3 effects [total ∆: 3.2 % points]>

20 percent of passengers make up two thirds of total fares.

>>> tf.the_rule()
'Rule 60/20 [total ∆: 3.0 % points]'

Inversely, 60 percent of passengers make up only 20 percent of fares.

The overview:

>>> tf.groups()
  Causes    Effects    Count    Mean    Stdev    Stdev/Mean    ∆ Mean
--------  ---------  -------  ------  -------  ------------  --------
    100%       100%      891  32.204   49.666           1.5
     20%        67%      185  99.565   76.537           0.8     +209%
     80%        33%      706  14.553    8.801           0.6      -55%

Findings

The standard deviation of 891 fares (in proportion to the standard) goes down almost by a factor of three if we segregate 185 passengers from them.

The fare price of 15 is way more representative for most passenger than the average of 32 of all passengers would suggest.

Despite the spike of the standard deviation of the vital few (the second row) in absolute terms, it dropped in proportion to the mean from 1.5 for all to 0.8 for the few.