Mr Andrew Bell

Research summary

My PhD looks at how context should be modelled and conceptually understood when using temporal (panel, repeated cross-sectional, or time-series cross-sectional) data. This includes:

• The difference between fixed and random effects, and the capabilities of the latter to model context explicitly.  The inclusion of ‘between effects’ in a random effects (multilevel) model allows this, as well as solving the problems of omitted variable bias and allowing complex, multilevel questions to be answered.
• The advantages and disadvantages of using hierarchical conditional models when using non-linear dependent variables, compared to single level marginal models such as General estimating equations (GEE), particularly in ...

Read more >

Links

•   Academia.edu profile
• Google Scholar profile

Selected publications

• Bell, AJD & Jones, K 2013, ‘Explaining Fixed Effects: Random Effects modelling of Time-Series Cross-Sectional and Panel Data’. Unknown.

Read more >

Recent publications

• Bell, A & Jones, K 2013, ‘The impossibility of separating age, period and cohort effects’. Social Science & Medicine.
• Bell, AJD & Jones, K 2013, ‘Current practice in the modelling of Age, Period and Cohort effects with panel data: a commentary’. Unknown.

Read more >

View complete publications list in the University of Bristol publications system