SQL for Data Science
This course is available only as a part of subscription plans
Training duration : 4 hours
Identify the contents of and relationships between tables in a database
Write queries to aggregate, filter and sort data in a table
Join multiple tables in a relational data
Perform mathematical operations between columns
Identify the need for a subquery in order to properly transform data
Write subqueries in the FROM and WHERE clauses
Write common table expressions
Answer business questions using a combination of methods listed
DIFFICULTY LEVEL: BEGINNER
Instructor Bio:
Mona Khalil
During the 4-hours of lectures and exercises, we will cover the following topics:
Lesson 1: Relational Databases and Foundational SQL (45 minutes)
Familiarize yourself with relational databases and the SQL syntax necessary to retrieve information from tables in a database. At the end of this lesson, you will be able to comfortably explore a database and retrieve filter, and sort information from a table.
Lesson 2: Combining Data From Multiple Tables and Columns (1 hour)
Practice joining tables in a database and aggregating that information to answer simple questions about the data in your database. You will be able to identify columns used for joining/combining tables, choose the correct method for joining tables, and perform simple mathematical calculations on your data.
Lesson 3: Layering your Transformation with Subqueries (1 hour)
Often, your data needs to be transformed in multiple steps to get it in the shape necessary for a specific task. By the end of this lesson, you’ll learn how to write subqueries, common table expressions, and simple window functions necessary to shape data in multiple steps.
Lesson 4: Transforming your Data for Analysis (45 minutes)
Put your skills to the test by answering some common business questions using a relational database. You will be able to leverage existing information in a database to create new columns, conditional statements, complex filters, and prepare a dataset for evaluation by stakeholders or more complex statistical analysis.
Being a core skill necessary to work with data effectively, the target audience is fairly wide and includes aspiring data analysts, business analysts, data scientists, and data engineers beginning to learn the core skills in their field.
Attendees should be familiar with the structure and manipulation of data in Excel/CSV files, but proficiency in Excel is not required.
CHECK OUT NEW AND FEATURED COURSES >>