Skip to main content

Poker test, 1000 four-digit numbers (2072 Magh)

Short answer

Using standard four‑digit probabilities, χ0² = 20.01. With 4 degrees of freedom and critical value 9.488 at α=0.05, since 20.01 > 9.488, the numbers are not independent.

A sequence of 1000 four digit numbers has been generated and an analysis indicates the following conditions. Based on poker test, are these numbers independent? (use α=0.05 and N = 4 is 9.488)

Combination DistributionsObserved Frequency
4 different Digits510
3 Like Digits40
4 Like Digits5
1 Pair410
2 Pair35

Setup. H0: the numbers are independent. H1: they are not. N = 1000, five patterns, df = 4, χ²0.05,4 = 9.488. The classes are rearranged in the usual order: 4 different, 1 pair, 2 pair, 3 like, 4 like.

Probabilities for four-digit numbers: 0.504, 0.432, 0.027, 0.036, 0.001.

PatternProbabilityOiEi(Oi − Ei)²/Ei
4 different digits0.5045105040.071
1 pair0.4324104321.120
2 pair0.02735272.370
3 like digits0.03640360.444
4 like digits0.0015116.000

χ0² = 20.01

Conclusion. Since 20.01 > 9.488, H0 is rejected; the numbers are not independent according to the poker test. Most of the value comes from "4 like digits" (5 observed against 1 expected, contribution 16.0).

Remark: if the small classes (3 like and 4 like, Ei = 37) were merged, df = 3 and the statistic would be 5.29 < 7.815, so independence would be accepted. With the critical value 9.488 given, the answer is to reject.

✓ Answer checked by Claude Opus 5.5 (high effort)

All Simulation and Modeling chapter 6 questions →