Scaling idea
Let
$$ X_n=\sum_{i=1}^n \varepsilon_i $$be a symmetric random walk with $\varepsilon_i=\pm 1$ equally likely.
For fixed $t>0$, consider the rescaled position
$$ \frac{1}{\sqrt n}X_{\lfloor nt\rfloor}. $$Limiting distribution
By the central limit theorem,
$$ \frac{1}{\sqrt n}X_{\lfloor nt\rfloor} \xrightarrow{d} N(0,t). $$The variance is
$$ \mathrm{Var}\left(\frac{1}{\sqrt n}X_{\lfloor nt\rfloor}\right)
\frac{\lfloor nt\rfloor}{n} \to t. $$
Why this matters
Brownian motion is the continuous-time, continuous-space limit of symmetric random walk under this joint time-space scaling.
Source Links
- STA447
- Phase 6 Atomic Reading Order
- Phase 6 — Brownian Motion 基础
- Previous atom: none
- Next atom: Brownian Motion