| Sumario: | A study was conducted to compare the aggregated spatial choice model and the popular ordinary multinomial logit model in light of previous findings that, under certain circumstances, the aggregated spatial choice model can outperform the other model both in explanatory power and predictive ability. The two models were compared using interprovincial migration microdata for the period 1990-91, obtained from the 1991 Canadian census. The analysis was limited to out-migrants from Ontario. Findings revealed that the multinomial logit model performed reasonable well, at least with the data used. Some practical advantages that can accrue from the use of the aggregated model, however, are outlined.
|
|---|