2

I strongly believe that this distribution does not belong to the exponential family:

$f(x;\theta) = \frac{\theta}{2}^{|x|}(1 - \theta)^{1-|x|}I_{\{-1, 0, 1\}}(x)$.

I have to write $f(x;\theta)$ as $a(\theta)b(x)\exp(c(\theta)d(x))$, I don't think this can be done. All I can do is write:

$f(x;\theta) = \exp(|x|\log(\theta) + (1-|x|)\log(1-\theta))I_{\{-1, 0, 1\}}(x)$.

jbowman
  • 38,614

1 Answers1

1

Since $$f(x;\theta) = \frac{\theta}{2}^{|x|}(1 - \theta)^{1-|x|}\mathbb{I}_{\{-1, 0, 1\}}(x)=\frac{1}{2}\left\{\frac{\theta}{(1-\theta)}\right\}^{|x|} \mathbb{I}_{\{-1, 0, 1\}}(x)$$the likelihood associated with a sample $(x_1,\ldots,x_n)$ is $$L(\theta|x_1,\ldots,x_n)=\prod_{i=1}^n f(x_i;\theta)= \frac{1}{2^n}\left\{\frac{\theta}{(1-\theta)}\right\}^{\sum_{i=1}^n|x_i|} $$ which factorises through the unidimensional $$\sum_{i=1}^n|x_i|$$ for all $n$'s. Thus an exponential family.

Xi'an
  • 105,342