Skip to main content

TL;DR

Information-Theoretic Sorting Lower Bound: A canonical quantitative trading interview question at intermediate difficulty. Commonly asked at Jane Street, Citadel, Two Sigma.

By Valenke Exam Prep Team·Last updated 2026-06-01
intermediateComplexity & Lower Bounds

Information-Theoretic Sorting Lower Bound

Asked at: Jane Street, Citadel, Two Sigma

Problem
Prove that any comparison-based sorting algorithm must perform at least ⌈log⁡2(n!)⌉\lceil \log_2(n!) \rceil comparisons in the worst case to sort nn distinct elements. Use this to establish the Ω(nlog⁡n)\Omega(n \log n) lower bound.

Ready to practice for the Valenke Finance Exam?

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

Start free practice →