A spatial autoregressive multinomial probit model for anticipating land-use change in Austin, Texas

作者: Yiyi Wang , Kara M. Kockelman , Paul Damien

DOI: 10.1007/S00168-013-0584-Y

关键词:

摘要: This paper develops an estimation strategy for and then applies a spatial autoregressive multinomial probit model to account both clustering cross-alternative correlation. Estimation is achieved using Bayesian techniques with Gibbs the generalized direct sampling (GDS). The applied analyze land development decisions undeveloped parcels over 6-year period in Austin, Texas. Results suggest that GDS useful method uncovering parameters whose draws may otherwise fail converge standard Metropolis-Hastings algorithms. results residential commercial/civic tends favor more regularly shaped smaller parcels, which be related parcel conversion costs aesthetics. Longer distances Austin’s central business district increase likelihood of development, while reducing office/industrial uses. Everything else constant, parcel’s nearest minor, major arterial roads are estimated office/industry uses, perhaps because such common less densely developed locations (as proxied by fewer arterials). As expected, added soil slope negatively associated but positively uses (perhaps due some steeper terrains offering view benefits). Estimates correlations use “utility” or attractiveness correlated commercial/civic, latter two exhibit negative Using inverse-distance weight matrix each closest 50 neighbors, autocorrelation coefficient 0.706, indicating marked pattern selected region.

参考文章(0)