Determinering av sol- og utsiktskvaliteter ved hjelp av geografiske informasjonssystemer og regresjonsanalyse

Abstract

The background for the thesis was a desire to find methods that determine prices of new homes and estimate how much the value of existing houses diminish when neighboring land is developed. The problem was as follows: "Is there a relationship between house prices in Trondheim and factors such as solar radiation, view and proximity to the center? What are the existing methods used by real estate brokers to appraise residential qualities? Can multivariate regression analysis be used to determine house prices? "

I solved these problems by using GIS to create an urban 3D model of Trondheim. The 3D model was used as the basis for calculating the sun and view conditions using a solar radiation and viewshed analysis respectively. When the GIS analysis was done, I used SPSS for statistical calculations and regression analysis. Based on the regression results, I conclude that there is a significant correlation between the purchase and view conditions of town houses in Trondheim. The methods I have used make it possible to calculate the solar radiation and viewshed and thus predict the purchase of houses by using a multivariate regression. The method also makes it possible to calculate the compensation for loss of views, due to for instance an increased building density.