I am supposed to show the constant of a Weibull Distribution, $k$, is a product of $\alpha$ and $\beta$.

I know $\alpha = \frac{1}{\theta}$ ($\theta$ being the probability of a success) for the Distribution becomes an exponential distribution when $\beta = 1$. But I'm still not sure how $k = \alpha\beta$. How do I show this? also, is there a specific name for this "process?"

enter image description here

  • $\begingroup$ Since $f$ is a probability distribution you know that $\int_\mathbb{R} f dx = 1$. Use this to solve for $k$. $\endgroup$ – Digitalis Apr 15 at 12:23

\begin{align} \int_{0}^{+ \infty} kx^{\beta - 1}e^{-\alpha x^{\beta}} dx &= 1 \\ \Rightarrow \int_{0}^{+ \infty} x^{\beta - 1}e^{-\alpha x^{\beta}} dx &= \frac{1}{k} \\ \end{align}

and using the transformation:

$$t = x^{\beta} \Rightarrow dt = \beta x^{\beta - 1}dx$$

we have:

\begin{align} \frac{1}{k} &= \int_{0}^{\infty} \frac{1}{\beta} e^{-\alpha t} dt \\ \Rightarrow \frac{\beta}{k} &= \int_{0}^{\infty} e^{-\alpha t} dt \\ \Rightarrow \frac{\beta}{k} &= \left [ -\frac{1}{\alpha} e^{-\alpha t} \right ]_{0}^{\infty} \\ \Rightarrow \frac{\beta}{k} &= \frac{1}{\alpha} \\ \Rightarrow k &= \alpha \beta \\ \end{align}


  • $\begingroup$ Don't think this kind of processes have a specific name, just note that often to answer this kind of questions, exploiting the fact that probability distribution integrated over their support sum up to $1$, instead of directly going on with math it can be useful to manipulate the quantity within the integral in order to obtain a new distribution. An example of this is the computation of the $k^{th}$ moment of a Gamma distribution. $\endgroup$ – Nicg Apr 15 at 13:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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