Mathwords logoMathwords

Bonferroni Inequalities — Definition, Formula & Example

The Bonferroni inequalities are a family of bounds on the probability that at least one of several events happens. You stop the inclusion-exclusion formula early, and the sum you have so far is guaranteed to be too high or too low depending on where you stopped.

Let A1,…,AnA_1, \dots, A_n be events and define Sj=∑1≤i1<⋯<ij≤nP(Ai1∩⋯∩Aij)S_j = \sum_{1 \le i_1 < \cdots < i_j \le n} P(A_{i_1} \cap \cdots \cap A_{i_j}). For each kk with 1≤k≤n1 \le k \le n, the partial sum ∑j=1k(−1)j−1Sj\sum_{j=1}^{k} (-1)^{j-1} S_j is an upper bound for P(⋃i=1nAi)P\left(\bigcup_{i=1}^n A_i\right) when kk is odd and a lower bound when kk is even. When k=nk = n the partial sum equals the probability exactly.

Key Formula

∑j=1k(−1)j−1Sj  ≤  P ⁣(⋃i=1nAi)  ≤  ∑j=1k+1(−1)j−1Sjfor even k\sum_{j=1}^{k} (-1)^{j-1} S_j \;\le\; P\!\left(\bigcup_{i=1}^{n} A_i\right) \;\le\; \sum_{j=1}^{k+1} (-1)^{j-1} S_j \quad \text{for even } k
Where:
  • AiA_i = The events, for i = 1 to n
  • SjS_j = Sum of the probabilities of all intersections of exactly j distinct events
  • kk = Number of inclusion-exclusion terms kept (even k gives the lower bound shown)
  • P(⋃Ai)P\left(\bigcup A_i\right) = Probability that at least one event occurs

Worked Example

Problem: You plan a three-day camping trip. The probability of rain on any one given day is 0.2. The probability of rain on any two given days is 0.05. The probability of rain on all three days is 0.01. Use the Bonferroni inequalities to bound the probability that it rains on at least one day.
Step 1: Compute the three sums. There are 3 single days, 3 pairs of days, and 1 triple.
S1=3(0.2)=0.6,S2=3(0.05)=0.15,S3=0.01S_1 = 3(0.2) = 0.6, \quad S_2 = 3(0.05) = 0.15, \quad S_3 = 0.01
Step 2: Keeping one term (odd) gives an upper bound. Keeping two terms (even) gives a lower bound.
0.6−0.15=0.45  ≤  P(rain at least once)  ≤  0.60.6 - 0.15 = 0.45 \;\le\; P(\text{rain at least once}) \;\le\; 0.6
Step 3: Keeping three terms (odd) gives a sharper upper bound. Because there are only three events, this third partial sum is the exact value.
0.6−0.15+0.01=0.460.6 - 0.15 + 0.01 = 0.46
Answer: The bounds give 0.45≤P≤0.460.45 \le P \le 0.46. With three events, the full inclusion-exclusion sum shows the probability is exactly 0.46.

Why It Matters

The first Bonferroni inequality, P(⋃Ai)≤∑P(Ai)P\left(\bigcup A_i\right) \le \sum P(A_i), is the basis of the Bonferroni correction in statistics. If you run mm hypothesis tests, each at significance level α/m\alpha/m, the chance of at least one false positive is at most α\alpha. This correction appears in medical trials, genetics studies, and A/B testing.

Common Mistakes

Mistake: Treating S1−S2S_1 - S_2 as an upper bound, or assuming every partial sum overestimates the probability.
Correction: The direction alternates. An odd number of terms gives an upper bound, and an even number gives a lower bound. In the camping example, S1−S2=0.45S_1 - S_2 = 0.45 falls below the true value 0.46.

Related Terms