SQL Tutorial: Time Series Analysis in SQL Server | Building dates

Want to learn more? Take the full course at https://learn.datacamp.com/courses/time-series-analysis-in-sql-server at your own pace. More than a video, you'll learn hands-on coding & quickly apply skills to your daily work. --- Welcome! In this course, we will look at a number of techniques for working with time series data using SQL Server. My name is Kevin Feasel. I am a Microsoft Data Platform MVP and am a co-founder at Envizage, a company specializing in data analysis and visualization. Throughout this course, we will use built-in functionality to solve a number of business problems with SQL Server. To start, we will learn how SQL Server handles the individual parts which make up a date and translating strings--for example, from CSV files--to dates. We will then take advantage of these functions to control the grain of our dates. This includes, for example, downsampling data to an hourly grain for analysis. We will then cover specific business cases around windows of data, including running totals, moving averages, and where date ranges overlap. In this first lesson, we will review some of the basics around date handling in SQL Server. You should already be familiar with the DATETIME and DATETIME2 data types. Let's review a few functions which use these date types to make your life easier. First, GETDATE() and GETUTCDATE() will return a date, in local or UTC time respectively, as a DATETIME data type. Similarly, SYSDATETIME() and SYSUTCDATETIME() return the current time--either local or UTC--as a DATETIME2 typed response. In both cases, we get the dates and times we expect. Suppose we don't need an entire date and time but only want a specific part. For a date like this one, there are a few built-in functions. We can get the year. Or the month. Or the day. This might not be enough to satisfy us, so we need to take it to the next level. The DATEPART() and DATENAME() functions give us much more control over our date parts. DATEPART() returns t