INVESTIGATING THE INTERACTION EFFECTS OF OFFICE RENT DETERMINANTS
||Öven, Vasif Attila; Dilek Pekdemir
||INVESTIGATING THE INTERACTION EFFECTS OF OFFICE RENT DETERMINANTS
||Book of Abstracts: 13th Annual European Real Estate Society Conference in Weimar, Germany
||The application of factorial design on product quality management is becoming increasingly more popular [1, 2, 3]. The essence of the numerical technique lies with its efficiency on determining both the influence of the main effects and their combined effects on optimising the desired qualities of an end product. Considering office rent is a function of a large number of parameters (office rent determinants) it may be prudent to apply the factorial design technique treating office rent as an end product whose quality is measured by the highest yield of rent Using the method, it is then possible to define the most influential parameters (main effects) as well as the parameter interactions (combined effects) dictating the highest office rent [4, 5]. In a previous work by the same authors , it was attempted to improve hedonic office rent prediction models that represent the large number of office rent determinants. The main criticism directed to the hedonic approach is its inability to assess the interrelation between the price attributes. However, the interrelated influence of certain attributes on rental value can be of a great value [4, 7]. “2n Factorial Design” approach is a useful statistical technique to investigate the influence of all possible combinations of the levels of the independent variables on the dependent variable. The aim of this paper is to determine the interrelated influence of certain attributes on rental value. Expanding the contract data used in the previous study , 2n two-level factorial design was conducted in order to examine the influence of interaction effects of the variables of the previous work on office rent in Istanbul.
||office rent determinants; experimental design; factorial design
Post discussion ...
These pages are best viewed with any standards compliant browser (e.g. Mozilla).