Time series analysis encompasses methods for examining time series data found in a wide variety of domains. Being equipped to work with time-series data is a crucial skill for data scientists. In this course, you'll learn to extract and visualize meaningful statistics from time series data. You'll apply several analysis methods to your project. Along the way, you'll learn to explore, analyze, and predict time series data.
You'll start by working with pandas' datetime and finding useful ways to extract data. Then you'll be introduced to correlation/autocorrelation time-series relationships and detecting anomalies. You'll learn about autoregressive (AR) models and Moving Average (MA) models for time series, and explore anomalies in detail. You'll also discover how to blend AR and MA models to build a robust ARMA model. You'll also grasp how to build time series forecasting models using ARIMA. Finally, you'll complete your own project on time series anomaly detection.
By the end of this practical tutorial, you'll have acquired the skills you need to perform time series analysis using Python.
Please note that this course assumes some prior knowledge of Python programming; a working knowledge of pandas and NumPy; and some experience working with data.
About the Author
Karen J. Yang has been a data engineer, an author, and a passionate computer science self-learner for 7 years. She has 6 years' experience in Python programming and big data processing. Her recent interests include cloud computing.
She holds a PhD in Political Science from Ohio State University and loves working with data to gather meaningful information by performing analysis and research. This interest led her to publish data analysis research papers on Inferential Data Analysis on Tooth Growth and Predicting Activity for Samsung SensorData. She is also a published author of the 'Apache Spark in 7 Days' course.
