Feng Shui & Vancouver real estate

The location of medical care centers (hospitals, clinics and hospices) and the location of funeral homes and crematoriums at displayed in Figure 1. Nearest Neighbor analysis was conducted in order to quantify which spatial locations would be ‘affected’ by their presence. The radius used in the analysis represents two blocks surrounding each point.

Figure1:

 

Preliminary rounds of GWR analysis determined that the best predictor of the model is then independent variable: ‘home square footage’. However, the R² value of the home square footage variable, yielded similar values (Table 1). Figure 2 also shows the result of the visual analysis.

Table 1: R Squared and Residual Square results for GWR analysis with various input variables

 Figure 2:

The addition of the Feng Shui variables was then added to the GWR model with ‘home square footage’. The addition of these variables did increase the predictability of the model, although only minimally (Table 1). Figure 3 compares the GWR results. 

Figure 3:

 

In order to reveal why the GWR analysis better predicted selling price with the inclusion of the Feng Shui attributes, the results were displayed on top of a chloropleth map. The Map indicates number of people with Chinese ethnic origin (in the 20% sample taken) in each dissemination area (Figure 4).

Figure 4:

 

 The influence of the unlucky 4 in an address, was determined by applying the T Test to tow categories of home sales: those with 4’s and those without. Under the assumption that the presence of a 4 should affect the sale price of a home, the results of the GWR (which did not include the 4 as a independent variable) should be different for the category containing 4’s then those homes that didn't’t. This was not the case, as the T-Test did not produce significant results (Table 2).

Table 2:

Lastly, a kernel surface was created to visually revel which areas of Vancouver had the highest density if negative Feng Shui locations.  The map is a result of adding the separate kernel layers together (Figure 5).

Figure 5: