Theory#

Introduction#

Basics#

Block pulse functions#

Define an \(m\)-set of block pulse functions as

\[\begin{split}\phi_i(t) = \begin{cases} 1 & , \ (i-1)h \leq t < ih \\ 0 & \text{otherwise} \end{cases} \qquad (i=1,\ldots,m)\end{split}\]

for \(t \in [0,T)\) and an interval width of \(h=\frac{T}{m}\). The BPFs are disjoint, i.e.

\[\phi_i(t)\phi_j(t) = \delta_{ij} \phi_i(t)\]

for \(i,j = 1, \ldots, m\) and \(\delta_{ij}\) the Kronecker delta and orthogonal, i.e.

\[\int\limits_0^T \phi_i(t) \phi_j(t) dt = h \delta_{ij}\]

for \(i,j=1,\ldots,m\).

Conditional expectation#

Brownian Motion#

Itô-Integral#

Stochastic integration operational matrix#

Solving stochastic Volterra integral equations using a stochastic operational matrix#

Examples#

Example 1#

Example 2#

Example 3#

Applications#