Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $$s_i \sim \chi(k_i, \lambda_i), i\in \{ 1, 2\}$$ be two non-central chi-squared random variables with $k_i$ degrees of freedom and $\lambda_i$ parameter of non-centrality I am wondering if $$p_{max} = max(s_1,s_2), p_{min} = min(s_1,s_2)$$ are also non-central chi squared random variables? What I quickly did was to implement it in MATLAB and I realized that both $$p_{max} , p_{min}$$ follow non central chi-squared distributions but I don't have a concrete proof of it. Also, if my assumption is true, what would be the relationships between $ k_i,\lambda_i$ and parameters for both of these $p_{max}, p_{min}$. If this is a known problem, please refer me to a text or similar because I have not found this problem solved anywhere.

share|improve this question

1 Answer 1

up vote 2 down vote accepted

Let $X$ and $Y$ be two independent random variables. Then: $$ F_{\max(X,Y)}(z) = \mathbb{P}\left(\max(X,Y) \leqslant z\right) = \mathbb{P}(X \leqslant z, Y \leqslant z) = \mathbb{P}(X \leqslant z) \mathbb{P}(Y \leqslant z) = F_X(z) F_Y(z) $$ Similarly: $$ 1- F_{\min(X,Y)}(z) = \mathbb{P}\left(\min(X,Y)> z\right) = \mathbb{P}(X > z, Y > z) = \mathbb{P}(X > z) \mathbb{P}(Y > z) = (1-F_X(z))(1-F_Y(z)) $$ Since the product of cumulative distribution functions of two non-central $\chi^2$ random variables does not have the functional form of the cumulative distribution of non-central $\chi^2$ random variable, your conclusion is premature. Neither $\min(X,Y)$ nor $\max(X,Y)$ follow a non-central $\chi^2$ distribution.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.