Katkat
Back

Maximum Profit from Dated Stock Trade Records

RampSoftware Engineer · Online Assessment
MediumSoftware EngineerOnline AssessmentCoding & Algorithms

Last updated

This question evaluates a candidate's ability to design an efficient single-pass algorithm for grouped time-series data, tracking per-symbol running minimums to compute maximum achievable profit. It tests array processing, hash-map grouping, and edge-case handling for unsorted records, commonly used in coding interviews to assess practical algorithmic problem-solving rather than pure theory.

Maximum Profit from Dated Stock Trade Records

You are given a log of stock trades. Each trade is a record [date, symbol, price]:

  • date — a calendar date as a string in "YYYY-MM-DD" format.
  • symbol — the stock's ticker, an uppercase string (e.g., "AAPL").
  • price — the trade price as a positive integer (the price in cents, so every value is exact).

The records may appear in any order — they are not sorted by date.

You want to make exactly one buy-then-sell transaction:

  • Buy one share of some stock on one day, then sell that same stock on a strictly later day.
  • The profit of a transaction is sell_price - buy_price.
  • The buy and sell must share the same symbol, and the sell date must be strictly after the buy date.

Return the maximum profit achievable from a single such transaction. If no transaction yields a positive profit (for example, fewer than two records exist for every symbol, or every symbol only declines over time), return 0.

Input / Output

  • Input: records, a list of [date, symbol, price] entries.
  • Output: a single integer — the maximum achievable profit, or 0 if none is positive.

Examples

Example 1:

records = [
 ["2024-01-03", "AAPL", 100],
 ["2024-01-01", "AAPL", 90],
 ["2024-01-02", "GOOG", 200],
 ["2024-01-04", "GOOG", 150]
]
Output: 10

Explanation: For AAPL, buy at 90 on 2024-01-01 and sell at 100 on 2024-01-03, giving a profit of 10. For GOOG, the price only falls (200 down to 150), so it offers no profit. The best single transaction is 10.

Example 2:

records = [
 ["2024-05-01", "TSLA", 300],
 ["2024-05-02", "TSLA", 250],
 ["2024-05-03", "TSLA", 200]
]
Output: 0

Explanation: TSLA only declines, so no profitable buy-then-sell exists.

Example 3:

records = [
 ["2024-07-10", "NVDA", 500]
]
Output: 0

Explanation: A single record cannot form a buy-then-sell pair.

Constraints

  • 0 <= len(records) <= 2 * 10^5.
  • Each date is a valid "YYYY-MM-DD" string; lexicographic string comparison matches chronological order.
  • 1 <= price <= 10^9.
  • Multiple records may share the same date and symbol; a buy and a sell on the same date do not form a valid transaction, because the sell date must be strictly later than the buy date.