|
City
|
In-sample
|
Out-of-sample
|
|---|
| |
Neigh.
|
City
|
Weeks
|
Neigh.
|
City
|
Weeks
|
|---|
|
Vitoria
|
0.63
|
0.91
|
217
|
0.49
|
0.30
|
34
|
|
GV
|
0.58
|
0.91
|
90
|
0.49
|
0.63
|
30
|
|
SL
|
0.51
|
0.86
|
86
|
0.33
|
0.70
|
16
|
- For each city, the model was fit to the data using the “in-sample” portion. The estimated parameters were then used to forecast the remaining data (“out-of-sample” portion). Goodness-of-fit was assessed as the Spearman’s correlation coefficient between the observed and predicted data at both the neighborhood and city levels for each portion of predicted data.
