Time Series Data Formats Made Easy
Sample eBook chapters (free): https://github.com/dataman-git/modern-time-series/blob/main/20240522beauty_TOC.pdf
eBook on Teachable.com: $22.50
https://drdataman.teachable.com/p/home
The print edition on Amazon.com: $65 https://a.co/d/25FVsMx
The Pandas data frame in most cases is sufficient to handle time series data. If the data is univariate time series, a Pandas series with a time index will work just fine. If the data is multivariate time series, a 2-dimensional Pandas data frame will work well. But how about a time series with probabilistic forecasts that have multiple values in each period? Figure (A) shows a multivariate case with the sales and temperature variables. The sales forecasts for each period have low, medium, and high possible values. Although Pandas may still be able to store this dataset, there are other more effective data formats designed to handle data with increasing complexity. A good understanding of the data formats will increase your productivity in time series modeling projects. This is the goal of the chapter.
