- 【Updated on May 12, 2025】 Integration of CiNii Dissertations and CiNii Books into CiNii Research
- Trial version of CiNii Research Knowledge Graph Search feature is available on CiNii Labs
- Suspension and deletion of data provided by Nikkei BP
- Regarding the recording of “Research Data” and “Evidence Data”
Panel Data Analysis with Distinction of the Direction and Path of Changes and the Status of Unchanges
-
- ARITA Shin
- University of Tokyo
Bibliographic Information
- Other Title
-
- 変化の向き・経路と非変化時の状態を区別したパネルデータ分析
- An Analysis of Income Changes Following Employment Status Changes
- 従業上の地位変化がもたらす所得変化を事例として
Search this article
Description
This paper aims to reveal the assumptions of the fixed effect model, which is widely used for analyzing panel data, and develop an alternate model that is able to distinguish the patterns of changes in independent variables. Although the fixed effect model cannot distinguish the direction and path of changes and the status of unchanges in independent variables mainly because it neglects the sequential order of panel data, in the real world, these elements in changes and unchanges may have unique effects on a dependent variable. By applying original parameterization to the first difference model, the author attempts to relax the assumptions of the fixed effect model and develop a model that can capture the unique effects of these elements on the premise that changes in independent variables directly cause changes in a dependent variable. The author uses this model to analyze income changes that follow changes in employment status, and theoretically investigates in what situations one should relax the assumptions of the fixed effect model and distinguish the direction and path of changes and the status of unchanges in independent variables.
Journal
-
- Sociological Theory and Methods
-
Sociological Theory and Methods 28 (1), 69-86, 2013
Japanese Association For Mathematical Sociology
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390282680143581952
-
- NII Article ID
- 130004685177
-
- ISSN
- 18816495
- 09131442
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- CiNii Articles
-
- Abstract License Flag
- Disallowed