# Bonferroni Inequality

Bonferroni's Inequality

## Bonferroni's Inequality

For $n$ events $A_1,A_2,\cdots, A_n$

 $$$$\label{bof} P\big(\cap_{i=1}^n A_i\big)\geq \sum_{i=1}^n P(A_i) -(n-1).$$$$

#### Proof

For $n =2$,

 $$\begin{equation*} P(A_1\cup A_2)= P(A_1)+P(A_2) -P(A_1\cap A_2) \end{equation*}$$

But $P(A_1\cup A_2)\leq 1$. Using this in above equation, we have

 $$\begin{eqnarray*} & &P(A_1\cup A_2)= P(A_1)+P(A_2) -P(A_1\cap A_2)\leq 1 \\ \implies & & P(A_1\cap A_2) \geq P(A_1) +P(A_2) -1. \end{eqnarray*}$$

Hence from above inequality, it is clear that the inequality \eqref{bof} is true for $n=2$.

Suppose the inequality \eqref{bof} is true for $n=r$.

 $$$$\label{bof2} P\big(\cap_{i=1}^r A_i\big)\geq \sum_{i=1}^r P(A_i) -(r-1).$$$$

Then

 $$\begin{eqnarray*} P\big(\cap_{i=1}^{r+1} A_i\big)& = & P\big(\cap_{i=1}^r A_i \cap A_{r+1}\big)\\ & \geq & P\big(\cap_{i=1}^r A_i\big) + P( A_{r+1}) -1\\ &\geq & \sum_{i=1}^r P(A_i) -(r-1)+ P(A_{r+1}) -1\\ & & \text{ (Using inequality \eqref{bof2})}\\ & \geq &\sum_{i=1}^{r+1} P(A_i) -(r+1-1). \end{eqnarray*}$$

Hence inequality \eqref{bof} is true for $n=r+1$. So by mathematical induction, the inequality \eqref{bof} is true for all $n$.

## Bonferroni's Inequality Example

If $P(A) = 0.9$ and $P(B) = 0.8$, show that $P(A \cap B)\geq 0.7$.

#### Solution

As $A\cup B \subseteq S$, we have $P(A\cup B) \leq P(S) = 1$.

But

 $$\begin{eqnarray*} P(A\cup B) &\leq& 1\\ P(A) + P(B)- P(A\cap B)& \leq & 1\\ P(A\cap B) &\geq & P(A) +P(B) -1 \end{eqnarray*}$$

Thus

 $$\begin{eqnarray*} P(A\cap B) &\geq & P(A) +P(B) -1\\ P(A\cap B) & \geq & 0.9 +0.8 -1\\ &\geq &0.7. \end{eqnarray*}$$

## Conclusion

In this tutorial, you learned about Bonferroni's Inequality and how to prove it.

To read more about the tutorials on Probability Theory refer the link Probability Theory. These tutorials will help you to understand basic concepts of probability and various important results of probability theory along with some numerical solved examples on probability theory.