Laserfiche WebLink
of Technology. Dr. Nelson's expertise for this project is in recent <br />work on hedonic modeling in semi-rural areas. <br /> <br />William Rabiega, Ph.D., Urban Planning, University of Oregon. <br />Dr. Rabiega's property value work has been in urbanized areas. <br /> <br />a. The Model of Hedonic Value Theory <br />The Model of Nedonic Value Theory is a research tool used to assign <br />dollar weights to a number of variables which have positive and negative <br />effects on property value as expressed in sale price. In results, the <br />variables are referred to as amenities (positive factors) and disamenities <br />(negative factors). In the model, values for each variable are compared <br />to the price at which the property sold to determine the significance <br />each factor had in calculating the final price. <br /> <br />Three concepts are important to an understanding of data interpretation <br />in using the Hedonic Model. The concepts are taken from statistics and <br />are not inherent to the Hedonic Model: <br /> <br /> · Si.gnificance (the likelihood a result does not occur by chance); <br /> · Beta weight (the size of the effect of a given independent variable <br /> on price); <br /> <br /> R2 (the amount of total variance in prices between houses explained <br /> by the variables). <br /> <br /> When the researchers construct the Hedonic Value Model, a list of variables <br /> which may have an effect on price of a ~'esidential property is assembled. <br /> As research progresses, it may become apparent that some variables are <br /> more important than others in determining sale price; some may not be <br /> important at ail. The variables which are determined to be important <br /> are those which have statistical significance. <br /> In an Hedonic Model, one is interested in relationships between phenomena. <br /> One hopes to establish a relationship between housing features and price. <br /> If the relationship is strong, it is significant. It is significant because <br /> it does not occur by chance. In statistics, significance of a relationship <br /> between two variables (e.g. a number of square feet in a house and <br /> sale price) is measured on a numerical scale of .00 to 1.00. The value <br /> is expressed as p=X. The closer the value is to .00, the greater the <br /> significance. This means the more likely the change in price occurs <br /> because of the variable, not because of chance. Variables with a <br /> significance value of (p=.05 or less) are considered to be significant <br /> and statistically reliable. Price is determined to be a function of the <br /> housing variables which show this level of significance. <br /> <br />6 <br /> <br /> <br />