Unfolding Beijing in a Hedonic Way

This is a public repository accompanying the paper

• Wei Lin, Zhentao Shi, Yishu Wang and Ting Hin Yan: “Unfolding Beijing in a Hedonic Way”.

for access of the data and R scripts. Please contact Yishu Wang (wangy@link.cuhk.edu.hk) if you have any questions about the code.

Code structure

Data:

• lianjia.RData: Transaction-level dataset of housing prices in Beijing from Lianjia.

Master Files:

• prediction.R: Main file for performing spatial prediction in Section 3.
• pred_seq.R: Main file for performing sequential forecast in Section 4.

KNN Tuning:

• KNN/plm.knn.R: Functions of partial linear k-Nearest Neighbor (KNN)
• KNN/SKNN.Tune.R: Tuning spatial KNN
• KNN/STKNN.Tune.R: Tuning spatial-temporal KNN
• KNN/SKNN_seq.Tune.R: Tuning sequential spatial KNN

NW Tuning:

• NW/plm.NW.R: Functions of partial linear Nadaraya-Watson (NW)
• NW/SNW.Tune.R: Tuning spatial NW
• NW/STNW.Tune.R: Tuning spatial-temporal NW
• NW/SNW_seq.Tune.R: Tuning sequential spatial NW

LPN Tuning:

• LPN/plm.localpoly.pred: Functions of partial linear Local Polynomial (LPN)
• LPN/SLPN.Tune.R: Tuning spatial LPN
• LPN/STLPN.Tune.R: Tuning spatial-temporal LPN
• LPN/SLPN_seq.Tune.R: Tuning sequential spatial LPN

RF Tuning:

• RF/RF.Tune.R: Tuning Random Forests (RF, spatial version)
• RF/RF_seq.Tune.R: Tuning Random Forests (RF, sequential version)

GBM Tuning:

• GBM/GBM.Tune.R: Tuning Gradient Boosting Machine (GBM, spatial version)
• GBM/GBM_seq.Tune.R: Tuning Gradient Boosting Machine (GBM, sequential version)

Visualization:

• GBMplot.R: Plotting Figure 3 "GBM prediction on coordinate raster"

Remark: If you want to save time from tuning parameters and trust our tuning results, you can directly run prediction.R and pred_seq.R to replicate our main results.