MTB > # Chapter 2 - predictive inference with discrete priors
MTB > # Example:  let p be the batting average of a 
MTB > # ballplayer -- either p=.2,.28 or .36 with probs
MTB > # .4, .5, .1, respectively.  Player gets 6 hits in 20 at-bats.
MTB > # What do you know about the number of hits of the player
MTB > # in 20 future at-bats?

MTB > exec 'p_disc_p'
 
INPUT 1 IF PROBABILITIES ARE IN 'PRIOR' OR
      2 IF PROBABILITIES ARE IN 'POST':
DATA> 2
 
INPUT NUMBER OF TRIALS:
DATA> 20
 
INPUT RANGE (LOW AND HIGH VALUES) FOR NUMBER OF SUCCESSES:
DATA> 0 20
 
 PREDICTIVE DISTRIBUTION OF NUMBER OF SUCCESSES:

 Row   SUCC       PRED

   1      0   0.004138
   2      1   0.023157
   3      2   0.064251
   4      3   0.118454
   5      4   0.163519
   6      5   0.179794
   7      6   0.162956
   8      7   0.124159
   9      8   0.080463
  10      9   0.044677
  11     10   0.021340
  12     11   0.008775
  13     12   0.003096
  14     13   0.000929
  15     14   0.000234
  16     15   0.000048
  17     16   0.000008
  18     17   0.000001
  19     18   0.000000
  20     19   0.000000
  21     20   0.000000