Kolmogorov-Smirnov one-sample test:
In theory, Monte Carlo studies rely on computers to generate large sets of random numbers. Particularly important are random variables representing the uniform pdf defined over the unit interval, fy(y) = 1,0 ≤ y ≤ 1. In practice, though, computers typically generate pseudorandom numbers, the latter being values produced systematically by sophisticated algorithms that presumably mimic "true" random variables. Below are one hundred pseudorandom numbers from a uniform pdf. Set up and test the appropriate goodness-of-fit hypothesis. Let α = 0.05.
|
.216
|
.673
|
.130
|
.587
|
.044
|
.501
|
.958
|
.415
|
.872
|
.329
|
|
.786
|
.243
|
.700
|
.157
|
.614
|
.071,
|
.528
|
.985
|
.442
|
.899
|
|
.356
|
.813
|
.270
|
.727
|
.184
|
.641
|
.098
|
.555
|
.012
|
.469
|
|
.926
|
.383
|
.840
|
.297
|
.754
|
.211
|
.668
|
.125
|
.582
|
.039
|
|
.496
|
.953
|
.410
|
.867
|
.324
|
.781
|
.238
|
.695
|
.152
|
.609
|
|
.066
|
.523
|
.980
|
.437
|
.894
|
.351
|
.808
|
.265
|
.722
|
.179
|
|
.636
|
.093
|
.550
|
.007
|
.464
|
.921
|
.378
|
.835
|
.292
|
.749
|
|
.206
|
.663
|
.120
|
.577
|
.034
|
.491
|
.948
|
.405
|
.862
|
.319
|
|
.776
|
.233
|
.690
|
.147
|
.604
|
.061
|
'.518
|
.975
|
.432
|
.889
|
|
.346
|
.803
|
.260
|
.717
|
.174
|
.613
|
.088
|
.545
|
.022
|
.459
|