Definition:Realization of Stochastic Process
Definition
Let $S$ be a stochastic process.
Let $T$ be a time series of observations of $S$ which has been acquired as $S$ evolves, according to the underlying probability distribution of $S$.
Then $T$ is referred to as a realization of $S$.
Thus we can regard the observation $z_t$ at some timestamp $t$, for example $t = 25$, as the realization of a random variable with probability mass function $\map p {z_t}$.
Similarly the observations $z_{t_1}$ and $z_{t_2}$ at times $t_1$ and $t_2$ can be regarded as the realizations of two random variables with joint probability mass function $\map p {z_{t_1} }$ and $\map p {z_{t_2} }$.
Similarly, the observations making an equispaced time series can be described by an $N$-dimensional random variable $\tuple {z_1, z_2, \dotsc, z_N}$ with associated probability mass function $\map p {z_1, z_2, \dotsc, z_N}$.
Examples
Batch Process
Consider a manufacturing plant which outputs a product at some variable rate.
Let the output generate a discrete time series $T$, obtained by accumulation.
The measurement of the volume of the output at the timestamps $T$ are the realizations of the underlying stochastic process governing that output.
Sources
- 1994: George E.P. Box, Gwilym M. Jenkins and Gregory C. Reinsel: Time Series Analysis: Forecasting and Control (3rd ed.) ... (previous) ... (next):
- Part $\text {I}$: Stochastic Models and their Forecasting:
- $2$: Autocorrelation Function and Spectrum of Stationary Processes:
- $2.1$ Autocorrelation Properties of Stationary Models:
- $2.1.1$ Time Series and Stochastic Processes: Stochastic Process
- $2.1$ Autocorrelation Properties of Stationary Models:
- $2$: Autocorrelation Function and Spectrum of Stationary Processes:
- Part $\text {I}$: Stochastic Models and their Forecasting: