Longest Stable Window

Finding the longest stable window in price data

This is a medium-difficulty coding problem that appears in quantitative trading interviews. It tests your ability to implement an efficient sliding-window algorithm that tracks the min and max of a moving range in real time.

The core challenge is maintaining both the maximum and minimum values within a window as you expand and contract it. Naive approaches (recalculating min/max from scratch at each step) will be too slow. Effective solutions use a data structure that supports efficient insertion, deletion, and range-query operations — such as a multiset, deque, or segment tree — to keep those bounds accessible in sublinear time. You'll also need to reason carefully about when to shrink the window and how to track the longest valid span you've seen.

• Sliding-window pattern and two-pointer logic
• Maintaining min/max efficiently over a dynamic range
• Choice of data structure (multiset vs. deque vs. custom tracking)
• Edge cases: empty input, threshold zero, uniform arrays

Strong solutions handle all test cases correctly, scale to large inputs without timeout, and explain the time/space trade-offs of their chosen approach.