SQL provides us with an easy way to automatically compute the moving average of a given column.
The moving average (also known as the rolling average, running average, moving mean (MM), or rolling mean) is a series of averages of different selections of the full data set. The moving average changes as we move through the data set. We can add the moving average to our SQL query result sets to see how it changes across the result set.
Continue readingWe can use SQL to create a running total of a given column. By this I mean, we can create a column that calculates the cumulative sum of a given column.
The running total/cumulative sum is the sum of all rows up until the current row. The result is that the sum increases (or decreases in the case of negative values) with each row returned in the result set. Each row’s value is added to the cumulative amount from all prior rows, so for any given row, we get the total sum for all rows up to that point – the “running total”.
Continue readingIn SQL, we can use the following method to compute the number of items within a given range when using a window function.
By this I mean we can use a frame clause in the window function to output how many items have a value that’s in a similar range to the value in the current row.
Continue readingWhen we use an OVER() clause to create a window function in SQL, we often use a PARTITION BY clause to partition the results. This can be handy if we want to do stuff like calculate subtotals.
But we can also use an empty OVER clause to calculate a grand total.
We might want to do this if we’re already using another OVER clause to calculate subtotals, but we also want a column to provide the grand total.
We can use the OVER() clause to create window functions in our SQL queries. A window function can be a useful tool that allows us to do things like compute moving averages, rank items, calculate cumulative sums, and much more.
Another common task for a window function might be to compute subtotals.
Continue readingSometimes when we run SQL queries we might want to group the results by a given column. A common way to do this is with the GROUP BY clause.
But sometimes the GROUP BY clause isn’t enough.
Sometimes we want the results presented in a way that the GROUP BY clause doesn’t cater for. The GROUP BY is perfect if we want to collapse groups of rows into a single row, so that we can get an aggregate value for each set of collapsed rows. But sometimes we don’t want this. Sometimes we want to see all of the rows, as well as the aggregate values.
This is where the PARTITION BY clause comes in.
When running SQL database queries, sometimes we need to use a window function in order to get the results we’re looking for. A window function is an aggregate-like function that enables partitioning and ordering of data within a result set.
The OVER clause is what enables us to create a window function.
The examples below demonstrate how we can incorporate the OVER clause in our SQL queries.
Window functions can be a useful tool when writing SQL queries. They allow us to include aggregate data across multiple rows without getting those pesky errors that sometimes occur when we try to use an aggregate function in the wrong way.
In this article, I aim to provide a simple overview of window functions and how they can be used to provide a more useful result set when running SQL queries.
Continue readingBehind every well-structured and organised SQL database lies a treasure trove of information called metadata. Metadata is data that provides information about other data. In relational database management systems, this metadata can typically include information like the names of databases, tables, columns, the data type of a column, access privileges, and much more.
The SQL standard provides a collection of tables/views that allow us to view metadata about the objects in the database. The SQL standard refers to this as the information schema and calls it INFORMATION_SCHEMA.
When learning SQL, a common mistake when checking for NULL values is to use an equality operator. In SQL, a NULL value is treated a bit differently to other values.
In SQL we use IS NULL instead of = NULL. Likewise, we use IS NOT NULL instead of <> NULL or != NULL to select those columns that don’t have a NULL value.