Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/207044 
Year of Publication: 
2019
Series/Report no.: 
Research Papers in Economics No. 12/19
Publisher: 
Universität Trier, Fachbereich IV - Volkswirtschaftslehre, Trier
Abstract: 
Spatial dynamic microsimulations allow for the multivariate analysis of complex socioeconomic systems with geographic segmentation. For this, a synthetic replica of the system as base population is stochastically projected into future periods. Thereby, the projection is based on micro-level transition probabilities. They need to accurately represent the characteristic dynamics of the system to allow for reliable simulation outcomes. In practice, transition probabilities are unknown and must be estimated from suitable survey data. This can be challenging when the characteristic dynamics vary locally. Survey data often lacks in regional detail due to confidentiality restrictions and limited sampling resources. In that case, transition probability estimates may misrepresent local dynamics as a result of insufficient local observations and coverage problems. The simulation process then fails to provide an authentic evolution. We present two transition probability estimation techniques that account for regional heterogeneity when the survey data lacks in regional detail. Using methods of constrained optimization and ex-post alignment, we show that local micro level transition dynamics can be accurately recovered from aggregated regional benchmarks. The techniques are compared in theory and subsequently tested in a simulation study.
Subjects: 
Constrained Maximum Likelihood
Logit Scaling
Spatiotemporal Modelling
Regional Benchmark
Document Type: 
Working Paper

Files in This Item:
File
Size
568.34 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.