Transaction Aggregation by Merchant and Hour

You are given a stream of payment transactions. Each transaction is represented as a string with the following comma-separated fields: transactionId, merchantId, amount, timestamp, status where status is one of {INITIATED, COMPLETED, FAILED, CANCELLED} and timestamp is given in the format "YYYY-MM-DD HH:MM:SS" (24-hour clock, UTC).

Write a function that receives a list of these transaction strings and an integer queryTimeUTC (Unix seconds since epoch). The function must return a JSON object (as a string) that aggregates only COMPLETED transactions whose timestamp is within the 24-hour window ending exactly at queryTimeUTC (inclusive on both ends). The aggregation must group first by merchantId and then by the hour of the day (0-23) in which the transaction occurred, and record the sum of amounts for each such bucket. The output JSON must contain nested objects: the top-level keys are merchantIds, each mapping to an object whose keys are hour strings "0" through "23" and whose values are the total summed amounts for that merchant in that hour. Merchants or hours that have no qualifying transactions may be omitted.