Expected value

Let $\mathbb{T}$ be a time scale. Let $X$ be a random variable with probability density function $f \colon \mathbb{T} \rightarrow \mathbb{R}$. The expected value of $X$ is given by $$\mathrm{E}_{\mathbb{T}}(X) = \dfrac{dC_f}{dz}(0),$$ where $C_f$ is the cumulant generating function of $f$.

Properties

Theorem: The following formula holds: $$\mathrm{E}_{\mathbb{T}}(X^k)=\displaystyle\int_0^{\infty} k! h_k(t,0)f(t) \Delta t.$$

Proof: █

Example

Theorem

Let $X$ have the uniform distribution on $[a,b] \cap \mathbb{T}$. Then, $$\mathrm{E}_{\mathbb{T}}(X) = \dfrac{h_2(\sigma(b),a)}{\sigma(b)-a}+a,$$ where $h_2$ denotes the delta hk and $\sigma$ denotes the forward jump.

Theorem

If $X$ is a random variable with the exponential distribution with parameter $\lambda$, then $$\mathrm{E}_{\mathbb{T}}(X)=\dfrac{1}{\lambda}.$$

References

Probability theory on time scales and applications to finance and inequalities by Thomas Matthews