Time series are observations on a variable ordered in time. They arise in many fields, including biology, telecommunications, physics, finance or economics.
One example from the world of finance is daily quotes of share indices, such as FTSE 100.
Time series analysis is a branch of statistics whose main aims are: (a) to find a model which provides a good description of the main features of the data, and (b) given the model and the data, to forecast and/or control the future evolution of the process. These two stages of analysis often require the development of novel procedures and algorithms which depend on the particular problem at hand.
Time series are used to analyse bodies of data over time, e.g., stock prices, environmental data, etc. Applications include such problems as predicting volcano eruptions.
- methodology and application of processes whose characteristics change "slowly" through time ("locally stationary" processes)
- Bayesian methods in time series analysis
- statistical filtering for general state-space models / inference in Hidden Markov models
- non-Gaussian time series, and in particular heavy-tailed time series often encountered in finance
- econometric time series
- spatio-temporal processes