Bigquery moving average

Question

Can anybody tell me how to calculate moving average in bigquery.

Here is what I need in mysql style.

SELECT T1.id, T1.value_column1, avg(T2.value_col         


        

Answer 1

For an updated and more efficient answer, https://stackoverflow.com/a/24943950/132438.

---

Check the new LAG() and LEAD() window functions. They allow you to traverse the result set, without the need for a self join.

https://developers.google.com/bigquery/docs/query-reference#windowfunctions

A different option with JOIN EACH (this can get too slow as an extremely large amount of data can get generated in the intermediate steps):

SELECT a.SensorId SensorId, a.Timestamp, AVG(b.Data) AS avg_prev_hour_load
FROM (
  SELECT * FROM [io_sensor_data.moscone_io13]
  WHERE SensorId = 'XBee_40670EB0/mic') a
JOIN EACH [io_sensor_data.moscone_io13] b
ON a.SensorId = b.SensorId
WHERE b.Timestamp BETWEEN (a.Timestamp - 3600000) AND a.Timestamp
GROUP BY SensorId, a.Timestamp;

(based on Joe Celko's SQL problems)