Skip to main content

TL;DR

Different ways a sequence of random variables can approach a limit: almost surely (a.s.), in probability (\( \xrightarrow{P} \)), in distribution (\( \xrightarrow{d} \)), and in \( L^p \) (mean). The implications are: a.s. \( \Rightarrow \) in probability \( \Rightarrow \) in distribution.

By Valenke Exam Prep Team·Last updated 2026-06-03

Modes of Convergence

Different ways a sequence of random variables can approach a limit: almost surely (a.s.), in probability (\( \xrightarrow{P} \)), in distribution (\( \xrightarrow{d} \)), and in \( L^p \) (mean). The implications are: a.s. \( \Rightarrow \) in probability \( \Rightarrow \) in distribution.

Why it matters for interviews

Understanding convergence modes is essential for rigorous probability theory: the LLN gives convergence in probability (weak) or a.s. (strong), while the CLT gives convergence in distribution. Interview questions test these distinctions.

Definition and Mathematical Foundation

Different ways a sequence of random variables can approach a limit: almost surely (a.s.), in probability (\( \xrightarrow{P} \)), in distribution (\( \xrightarrow{d} \)), and in \( L^p \) (mean). The implications are: a.s. \( \Rightarrow \) in probability \( \Rightarrow \) in distribution.

Application in Quantitative Finance

Understanding convergence modes is essential for rigorous probability theory: the LLN gives convergence in probability (weak) or a.s. (strong), while the CLT gives convergence in distribution. Interview questions test these distinctions.

Related Terms

Ready to practice for the Quant Trading Interview?

Adaptive practice powered by Item Response Theory targets your weak areas. Start with 3 free sessions.

Start free practice →

Frequently Asked Questions

What is the difference between convergence in probability and almost sure convergence?▼
In probability: for each \( \epsilon \), \( P(|X_n - X| > \epsilon) \to 0 \). Almost sure: \( P(X_n \to X) = 1 \). A.s. is stronger -- it says convergence happens for almost every outcome, not just that deviations become unlikely.
Why does convergence in distribution not imply convergence in probability?▼
Convergence in distribution only constrains the CDFs, not the joint behavior. Two variables can have the same limiting distribution without being close to each other. Example: \( X_n = X \) and \( X_n = -X \) both converge in distribution to X.
What is the continuous mapping theorem?▼
If \( X_n \xrightarrow{d} X \) and g is continuous, then \( g(X_n) \xrightarrow{d} g(X) \). This is used to derive the distributions of transformed statistics.