Project "Models and Methods of Demographic Sequence Analysis"

Sequence Analysis is one of the most advanced methods of analysis of demographic data. It was founded in the middle of the last century, but it was seriously developed only with the advent of data mining methods. Demographers, together with specialists in the field of information technology, aim to study the sequences of occurances of different events in the lives of Russians. Hidden patterns in people's behaviour are analyzed and new approaches to the prediction of the demographic trajectories are being developed. 

Aims and Objectives
The group has set itself two main objectives: the practical (the answers to demographic issues) and technical and mathematical (development and adaption of models and methods to answer demographic questions).

Practical problems include important areas of research in life paths, such as:

1. Analysis of transition into adulthood (the sequence of starting sociodemographic events).
2. Analysis of matrimonial and reproductive biographies (the sequence of marriages, partnerships and births).
3. Analysis of starting an independent life (a sequence of events such as: separation from parents, first work and receiving the highest level of education).
4. Analysis of movements in the course of life (a sequence of migrations by reasons and types of settlements).
5. Analysis of the life paths of men with and without experience of service in the army (the starting sequence of socio-demographic events in the life of men).

Technical, mathematical and methodological problems:

1. The study of methods and approaches for the analysis of sequences that already exist at the intersection of computer science and demography.
2. The development of easy-to-use tools for the analysis of sequences for demographers and sociologists.
3. The development of a method to distinguish generations based on data about sequences of socio-demographic events occurrences.
4. The development of a method to transfer the data of the format of events (date for each respondent) to the format of statuses (character codes for each respondent for each month of life within the specified period).
5. Creating a unified mathematical model for the analysis of demographic sequences at the junction of applied algebra, combinatorics (Galois connection, closure operators and the theory of lattices) and statistics (frequency analysis).

Dates of the project: January 2016 - January 2018 (2 years).

Expected scientific results of the research: scientific publications, presentations at conferences, interdisciplinary discussion, training of students in different aspects of sequence analysis both within the framework of the group and within the framework of the disciplines taught by the head of the group.

The results obtained during the research will help to understand how the life paths of Russians are structured. Within the research and study group, there are several major thematic blocks, such as: the transition to the adult life of Russians and, separately, men who served and did not serve in the army; marriage-partner behavior of generations; separation from parents in the context of related events; mobility of Russians throughout life.

Separate and no less important research results will be mathematical models and methods based on data mining and machine learning, as well as software implementations of these methods. In particular, models and methods based on closed frequent sequences, classifiers based on emergent patterns, prefix trees based on pattern structures.

The development of advanced methods for analyzing life events is already being introduced by the head of the group to the courses taught at the Higher School of Economics, for example, Methods of Demographic Analysis and Forecast, Theory and Methods of the Life-Course Approach, Research seminar "Population and Development". Also, joint work on the development of methods for researchers from different faculties strengthen interdisciplinary collaboration at the university and promotes the search for new and unique approaches to the analysis.

Form of presentation of the project results: scientific articles, reports at conferences, publications, sections of training courses.