Time Series

Overview

A time series is a series of measured values with a timestamp attached to each measurement. For example, a series of equity asset prices may be taken at the close of the market every day. Then you would have a series of data points, each with a date and price.

A Random Walk

copy

Modeling

• Definitions and Theorems - gives a mathematical description of time series and some important theorems
• Models - is the process of building a mathematical model to describe a time series, often for the purpose of forecasting.

Data Preparation

Outlines some of the issues surrounding preparing a dataset for a time series analysis.

• Building a Single Dataset : Time series data often comes from multiple datasets. Prior to analyzing the data, it is often necessary to combine the data into a single set.
• Data Cleansing
• Time Series Features/Transformations : Most time series datasets require a set of data transformations applied in order to extract features of the data that can then be modeled. The process of feature extraction is a core process of statistical/analysis and machine learning. However, due to the unique nature of time series data, the process of extracting features from a time series is unique to itself. Sample Features include:
  
  
  • Differences including returns
  • Volatility
  • Moving Averages and Filters