Research subjects
Statistics is the science of methods for extracting knowledge from data. Research in statistics develops strategies for efficient data collection, flexible probability models for analysing varying data types and complex relationships, and how the associated uncertainty should be quantified and communicated.
Methods and models need to be continuously developed to handle data of all kinds, ranging from sparse to large-scale data, in the form of both traditional numerical and categorical data as well as text, audio, and image data. Also, attention must be given to whether observations represent a cross-section at a given point in time, are arranged in time series, or structured spatially or in a network. The way in which observations are arranged often implies dependencies due to connections in time or space, or interactions and relationships between data points, which requires appropriate model adaptations.
Data may also be more or less structured depending on whether collection has been carried out through sample surveys, observational studies, or experiments, or whether data has been obtained from registers, through web scraping, via technical sensors, and so on. For research, this presents different challenges for data quality and considerations of the requirements needed for causal conclusions to be drawn.
Through methodological development in inference, control of various sources of error, and the development of specialised probability models, research ensures that valid conclusions can be drawn about underlying mechanisms based on observed data. Methods for accurate predictions of new data and for decision-making under uncertainty with minimised risk are also important components of statistics.
Research in statistics involves, in addition to statistical theory, mathematics, linear algebra, and computationally intensive methods for numerical optimisation. Statistical programming for efficient data management and visualisation, as well as the use of algorithms for simulations, makes advanced computational methods increasingly central.
Statistical methods are used across all empirical fields of science, and statistics researchers therefore often work within specialised subfields such as econometrics, psychometrics, demography, and biostatistics. Statistics is also fundamental to artificial intelligence and machine learning, areas that in recent years have become increasingly important for statistical research.