Time series data is omnipresent in the field of Data Science. Whether it is analyzing business trends, forecasting company revenue or exploring customer behavior, every data scientist is likely to encounter time series data at some point during their work. To get you started on working with time series data, this course will provide practical knowledge on visualizing time series data using Python.
You will learn how to leverage basic plottings tools in Python, and how to annotate and personalize your time series plots. By the end of this chapter, you will be able to take any static dataset and produce compelling plots of your data.Welcome to the course!50 xpLoad your time series data100 xpTest whether your data is of the correct type100 xpPlot your first time series50 xpYour first plot!100 xpSpecify plot styles100 xpDisplay and label plots100 xpCustomize your time series plot50 xpSubset time series data100 xpAdd vertical and horizontal markers100 xpAdd shaded regions to your plot100 xp
Summary Statistics and Diagnostics
In this chapter, you will gain a deeper understanding of your time series data by computing summary statistics and plotting aggregated views of your data.Clean your time series data50 xpFind missing values100 xpHandle missing values100 xpPlot aggregates of your data50 xpDisplay rolling averages100 xpDisplay aggregated values100 xpSummarize the values in your time series data50 xpCompute numerical summaries100 xpBoxplots and Histograms100 xpDensity plots100 xp
Seasonality, Trend and Noise
You will go beyond summary statistics by learning about autocorrelation and partial autocorrelation plots. You will also learn how to automatically detect seasonality, trend and noise in your time series data.Autocorrelation and Partial autocorrelation50 xpAutocorrelation in time series data100 xpInterpret autocorrelation plots50 xpPartial autocorrelation in time series data100 xpInterpret partial autocorrelation plots50 xpSeasonality, trend and noise in time series data50 xpTime series decomposition100 xpPlot individual components100 xpA quick review50 xpVisualize the airline dataset100 xpAnalyze the airline dataset100 xpTime series decomposition of the airline dataset100 xp
Work with Multiple Time Series
In the field of Data Science, it is common to be involved in projects where multiple time series need to be studied simultaneously. In this chapter, we will show you how to plot multiple time series at once, and how to discover and describe relationships between multiple time series.Working with more than one time series50 xpLoad multiple time series100 xpVisualize multiple time series100 xpStatistical summaries of multiple time series50 xpPlot multiple time series50 xpDefine the color palette of your plots100 xpAdd summary statistics to your time series plot100 xpPlot your time series on individual plots100 xpFind relationships between multiple time series50 xpCompute correlations between time series100 xpVisualize correlation matrices100 xpClustered heatmaps100 xp
Case Study: Unemployment Rate
This chapter will give you a chance to practice all the concepts covered in the course. You will visualize the unemployment rate in the US from 2000 to 2010.Apply your knowledge to a new dataset50 xpExplore the Jobs dataset100 xpDescribe time series data with boxplots100 xpBeyond summary statistics50 xpPlot all the time series in your dataset100 xpAnnotate significant events in time series data100 xpPlot monthly and yearly trends100 xpDecompose time series data50 xpApply time series decomposition to your dataset100 xpVisualize the seasonality of multiple time series100 xpCompute correlations between time series50 xpCorrelations between multiple time series100 xpInterpret correlations50 xpCongratulations!50 xp
In the following tracksTime Series with Python
DatasetsInventions and Scientific DiscoveriesCO2 levels in Mauai HawaiiAirline passengersProduction of meat in USAUnemployment rate in USA
Thomas VincentSee More
Head of Data Science at Getty Images
Thomas is an experienced statistician and programmer who is passionate about developing tools and pipelines to discover and retrieve underlying phenomenons and patterns in modern-day datasets. He enjoys applying his statistical skills to solve practical problems and blogs about his analyses at tlfvincent.github.io.