Rental Prices for Apartments in the province of Zurich (cont.) - - PowerPoint PPT Presentation

rental prices for apartments in the province of zurich
SMART_READER_LITE
LIVE PREVIEW

Rental Prices for Apartments in the province of Zurich (cont.) - - PowerPoint PPT Presentation

Non-Spatial Spatial Rental Prices for Apartments in the province of Zurich (cont.) Assignment 3 for Spatial Statistics (STAT 946) Adrian Waddell University of Waterloo December 2, 2008 Adrian Waddell (University of Waterloo) GRASS December


slide-1
SLIDE 1

Non-Spatial Spatial

Rental Prices for Apartments in the province of Zurich (cont.)

Assignment 3 for Spatial Statistics (STAT 946) Adrian Waddell

University of Waterloo

December 2, 2008

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 1 / 15

slide-2
SLIDE 2

Non-Spatial Spatial

Idea

Have information of n = 3088 apartments for rent available (on Friday October 3, 2008, homegate.ch) Trying to get a spatial and non-spatial model for the rental price. Idea is to find good offers. (i.e. look at residuals) Idea is inspired by Tobias Roithner.

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 2 / 15

slide-3
SLIDE 3

Non-Spatial Spatial

Model

Use Model log(price) = m(·) + e(s) e(s) = f(s) + ǫ non-spatial trend: m(area, nrRooms, ...) is chosen to be a linear model → variable selection spatial trend: e(s), model Variogram, Kriege residuals: ǫ

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 3 / 15

slide-4
SLIDE 4

Non-Spatial Spatial

Data

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 4 / 15

slide-5
SLIDE 5

Non-Spatial Spatial

Data

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 5 / 15

slide-6
SLIDE 6

Non-Spatial Spatial

Data

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 6 / 15

slide-7
SLIDE 7

Non-Spatial Spatial

Data

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 7 / 15

slide-8
SLIDE 8

Non-Spatial Spatial

Living Area

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 8 / 15

slide-9
SLIDE 9

Non-Spatial Spatial

Prediction Grid

In R: with spsample(ploy,2000,type = "regular")

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 9 / 15

slide-10
SLIDE 10

Non-Spatial Spatial

Inverse Weights

exp(e(s))

230000 240000 250000 260000 270000 280000 680000 690000 700000 710000 0.5 1.0 1.5 2.0 2.5

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 10 / 15

slide-11
SLIDE 11

Non-Spatial Spatial

Inverse Weights 0.7 < exp(e(s)) < 1.6

exp(e(s))

230000 240000 250000 260000 270000 280000 680000 690000 700000 710000 0.8 1.0 1.2 1.4 1.6

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 11 / 15

slide-12
SLIDE 12

Non-Spatial Spatial

Variogram: Mattern

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 12 / 15

slide-13
SLIDE 13

Non-Spatial Spatial

Ordinary Kriging

exp(e(s))

230000 240000 250000 260000 270000 280000 680000 690000 700000 710000 0.5 1.0 1.5 2.0

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 13 / 15

slide-14
SLIDE 14

Non-Spatial Spatial

Ordinary Kriging 0.7 < epx(e(s)) < 1.6

exp(e(s))

230000 240000 250000 260000 270000 280000 680000 690000 700000 710000 0.6 0.8 1.0 1.2 1.4 1.6

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 14 / 15

slide-15
SLIDE 15

Non-Spatial Spatial

End

THANK YOU

Adrian Waddell (University of Waterloo) GRASS December 2, 2008 15 / 15