Eres : Digital Library : Works

Paper eres2006_119:
THETA MODEL FORECASTS REAL ESTATE VALUES

id eres2006_119
authors Assimakopoulos, V.; A. Litsa; E. Pagourtzi
year 2006
title THETA MODEL FORECASTS REAL ESTATE VALUES
source Book of Abstracts: 13th Annual European Real Estate Society Conference in Weimar, Germany
summary The paper compares time series techniques that were used to forecast housing prices in UK, both at each region separately and as a whole. The tool used to provide forecasts is Theta Forecaster, a forecasting information system designed and developed at the Forecasting Systems Unit of National Technical University of Athens. This tool includes the Theta Model and several others, well-established forecasting methods. Theta Forecaster allows the combination of two or more techniques, which in many cases produces better forecasts than using a stand-alone method. The main feature of the Theta Model is that it applies different techniques to deal with sort-term and long-term forecasts and allows giving different weights in the sort and long-term components. The results show that Theta is always among the best forecasting methods and in many cases the most accurate one. The time series data used for forecasting is provided from the Halifax House Price Index, the UK's longest running monthly house price series covering the whole country from January 1983. The UK Index is derived from the mortgage data of the country's largest mortgage lender, which provides a robust and representative sample of the entire UK market. There are a number of national indices covering different categories of buyers (all, first-time buyers and home-movers) and houses, concerning the age (all, new and existing) and the type (all, detached, flats etc) of property. Regional indices for the 12 standard planning regions of the UK are available on a quarterly basis, while nationwide indices are produced monthly as well.
keywords Theta Model; real estate valuation; Theta forecaster; forecasting information systems
series ERES:conference
type normal paper
discussion No discussions. Post discussion ...
ratings
session 1-C
last changed 2008/11/01 09:47
HOMELOGIN (you are user _anon_893819 from group guest) Powered by SciX Open Publishing Services 1.002