1 Axioms of Probability Theory

1.1 Manipulation of Sets

Sets in R are represents as vectors.

For example, consider three such vectors generated at random from a set of integers.

Omega = 1:24
A = sample(Omega, 6)
B = sample(Omega, 9)

cat(Omega)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

cat(sort(A))

4 7 9 12 19 20

cat(sort(B))

5 9 14 17 19 20 22 23 24

The union of A and B

U = union(A, B)
cat(sort(U))

4 5 7 9 12 14 17 19 20 22 23 24

The intersection of A and B

U = intersect(A, B)
cat(sort(U))

9 19 20

The (set) difference of A minus B

U = setdiff(A, B)
cat(sort(U))

4 7 12

The symmetric difference of A and B

U = union(setdiff(A, B), setdiff(B, A))
cat(sort(U))

4 5 7 12 14 17 22 23 24

The following checks whether A and B are the same sets

setequal(A, B)

[1] FALSE

The following checks whether identifies the elements in A that are also in B

is.element(A, B)

[1]  TRUE FALSE FALSE  TRUE FALSE  TRUE

1.2 Venn and Euler diagrams

Venn diagrams are provide schematic visual representations of sets and their relationships. Euler diagrams are similar to Venn diagrams where only existing instersections are drawn.

require(eulerr)

The sample space is the set of integers between 1 and N.

N = 100
Omega = 1:N

The following are the elements of Omega that are divisible by 2, 3, and 5, respectively.

D2 = (Omega %% 2 == 0)
D3 = (Omega %% 3 == 0)
D5 = (Omega %% 5 == 0)

fit = euler(cbind(D2, D3, D5))

plot(
  fit, 
  labels = NULL, #' c("Divisible by 2", "Divisible by 3", "Divisible by 5")
  quantities = TRUE
)

Note that the areas of the various sets and subsets (their intersection) are not in proportion to their cardinality. (The cardinality is indicated on the plot.) Although it would be ideal if this were the case, it is not possible to draw such a diagram in general, at least not with discs. This is quantified as follows (see the package manual for details).

print(fit)

         original fitted residuals regionError
D2             27 27.057    -0.057       0.012
D3             14 14.094    -0.094       0.007
D5              7  7.193    -0.193       0.006
D2&D3          13 12.793     0.207       0.003
D2&D5           7  6.614     0.386       0.002
D3&D5           3  0.414     2.586       0.035
D2&D3&D5        3  3.567    -0.567       0.009

diagError: 0.035 
stress:    0.006 

However, we can use ellipses instead of discs to achieve a more faithful diagram. (The package provides a way to do that.)