Special Session 1

Advanced Statistical Methods in Demographical Research

 

Organizers: Arezoo Bagheri1, Mahsa Saadati2, Sohel Rana3
1,2National Population Studies & Comprehensive Management Institute, Tehran, Iran
3Department of Applied Statistics, East West University, Dhaka-1212, Bangladesh
Email: abagheri_000@yahoo.commahsa.saadati@gmail.comsrana@ewubd.edu

Aim: Socio-economic developments are related to four demographic dimensions, fertility, migration, mortality, and marriage. To achieve or sustain these developments, it is a necessity for countries around the world to study demographic data by new approaches. So, acquainting demographic researchers to these methods and the necessity of their applications are the first concern of this proposed Special Session. Three main topics of this session are considered as new developments in sampling procedures, in detection of outliers, and in classification of socio-demographic data. This session presents an advanced sampling method to analyze respondent driven samples. Another vital issue in this session which affects the validity of analysis is to detect anomalous data precisely in demographic surveys by defining a convenient robust detection method. Since demographic data mostly collected in national surveys and considered as big data, advanced statistical methods such as data mining should be applied to study them. Classification is one of the most feasible methods in data mining which recently population researchers are using them to investigate fertility behaviors, and migration. Classification and regression trees (CART) is also illustrated in this session. Event history analysis is a collection of statistical methods for the analysis of longitudinal data on the occurrence and timing of events. Event history analysis has become quite popular in medical and demographic fields with applications to such diverse events as birth, death, divorce, job change, and migration. Conventional survival models such as non-parametric and semi-parametric methods are not statistically valid in many situations when assumptions are not met. Researchers may also face to the correlated time events and the conventional survival methods do not consider these correlations. Therefore, variance estimations of the coefficients will be inflated and resulted models will be invalid. So, the other concern of this proposed special session are considered in the main two topics as application of parametric, and recurrent event survival models in demographical events such as the first marriage age, and birth.

Topics:
1- Respondent driven sampling in social harm studies
2- Robust generalized linear regression; Application in demographical studies
3- Data mining for Classification of demographical data
4- Parametric Survival Models in demographical researches
5- Recurrent event models in demographical event history data

Bulletin Board

Currently:

The Conference Guide is online.

The paper submission deadline has expired. Please choose a future conference to submit your paper.


Plenary Speakers

WSEAS Main Site

Publication Ethics and Malpractice Statement