运行 BigQuery 过去 30 天的总和

Question

我正在尝试在 BigQuery 中的 Google 商品商店 public 数据集上获取以下查询：

Date

Number of distinct users

Running sum of the number of distinct users in the last 30 days

例如（为简单起见，我在示例中使用了 3 天）：

  date        distinct_users     distinct_users_3days
15/07/2018          8                    15
14/07/2018          2                    12
13/07/2018          5                    20
12/07/2018          5                    15
11/07/2018         10                    10
   ...

这是我当前的 SQL 代码，它获取前两列，但我不知道如何获取运行总和：

SELECT
  date,
  COUNT(DISTINCT(fullVisitorId)) as daily_active_user
FROM
  `bigquery-public-data.google_analytics_sample.ga_sessions_2017*`
WHERE
  _table_suffix BETWEEN "0101"
  AND "0715"
GROUP BY
  date

感谢任何帮助！ :)

Answer 1

请尝试以下查询 3 天 (SQL server 2014 )-:

SELECT date,COUNT(DISTINCT(fullVisitorId)) as daily_active_user,sum(COUNT(DISTINCT(fullVisitorId))) over (PARTITION BY null ORDER BY date desc ROWS 
BETWEEN CURRENT ROW AND 2 FOLLOWING) AS distinct_users_3days FROM YOUR_TABLE_NAME WHERE _table_suffix BETWEEN '0101' AND '715' GROUP BY date

30 天-:

SELECT 
date,COUNT(DISTINCT(fullVisitorId)) as daily_active_user,
sum(COUNT(DISTINCT(fullVisitorId))) over (PARTITION BY null ORDER BY date desc ROWS 
BETWEEN CURRENT ROW AND 29 FOLLOWING) AS distinct_users_3days 
FROM YOUR_TABLE_NAME 
WHERE _table_suffix 
BETWEEN '0101' AND '715' 
GROUP BY date

Answer 2

我设法找到了问题的答案，所以我想与将来可能遇到此问题的其他人分享。

SQL代码是：

SELECT
  date,
  COUNT(DISTINCT(fullVisitorId)) as daily_active_user,
  SUM(count(Distinct(fullVisitorId))) OVER (ORDER BY date ROWS BETWEEN 29 PRECEDING AND CURRENT ROW) AS monthly_active_user
FROM
  `bigquery-public-data.google_analytics_sample.ga_sessions_2017*`,
  unnest(hits) as h
WHERE
  _table_suffix BETWEEN "0101" AND "0715"
GROUP BY
  date

这给出了一个列，该列汇总了 30 天内的不同用户 window。

运行 BigQuery 过去 30 天的总和

Running Sum for the last 30 days on BigQuery

sql

cumulative-sum

google-bigquery