Time Series with Zillow’s Luminaire — Part I Data Exploration

I sit at my local beach at midnight, enjoying the big moon casting sparkling lights on the ocean. The wavy moonlight path looks to me like a time series path, not always smooth but traceable. It shows me tranquility and serendipity.

I wrote a few articles on time series forecasting and anomaly detection. The Luminaire by the Zillow Tech Hub is the next one to that I want to write an in-depth introduction. Are you doing time series forecasting and outlier detection now or shortly? After reading this article, you will be running your time series model comfortably with Luminaire. Read this article and the following series: Anomaly Detection for Time Series, Business Forecasting with Facebook’s “Prophet”, “A Technical Guide on RNN/LSTM/GRU for Stock Price Prediction”, and “Kalman Filter Explained!”.

Change Points Are Challenging in a Univariate Time Series

You typically deal with a daily, monthly, or quarterly time series (and high-frequency in IoT or stock tick data). If the number of data points is not enough to train a model, someone may suggest taking a longer time range. However, a longer time series also brings a new challenge because it contains more change points. A change point is one in that the data trend shifts abruptly. This happens in almost all time series such as finance…