Fundamentals of time series data

Time Series Analysis in Power BI

Kevin Barlow

Data Analytics Professional

Course pre-requisites

This course requires intermediate DAX functions throughout the exercises, including:

CALCULATE()
IF()
SUMMARIZE()

Why is it important?

Every dataset has time as a dimension! We can learn some very interesting trends from analyzing data over time, and we need a strong toolkit to do so.

How has my data changed over time?
What should my data show based on history?
What will my data show in the future?

Why is time series data important

Defining time series data

Time Series Data is a set of observations (i.e. data points) that have been collected about the same subject over a period of time.

Two key components

Span: the total time period we have data
Interval: the amount of time in between each observation.

Time Series Components

Tracking changes over time

Secular Variation

Seasonal Variation

Cyclical Variation

Random (Irregular) Variation

Random Variation Example

Short term time series analysis

A short-term time span encompasses a timeframe typically less than one year. With these, we want to understand what our data is showing right now.

Common Analyses

Last x periods (e.g., Last 90 days, Last 2 quarters)
Period to date (e.g., Year to date, Month to Date)

Example

Short Term Analyses

Long term time series analysis

Long term analyses encompass a timeframe longer than one year. We are trying to understand historical data and its relationship to current data.

Common Analyses:

Year over Year
Month over Month
Same period last year

Example Calculating Year over Year (YoY) sales performance for the month of January.

YoY Analysis

Forecasting the future

Future analyses, such as forecasting, look at historical data and project it into the future. These kinds of analyses allow us to predict where we will be and make decisions based on that.

Common Analyses

Regressions (e.g. linear)
Point in time estimations
Machine Learning

Example

Forecasting stock trade volume several months into the future.

Forecasting Example

Let's practice!

Time Series Analysis in Power BI