The Advanced forecasting information system PYTHIA An application in real estate time series
Author: Pagourtzi, Elli ; Makridakis, Spyros ; Assimakopoulos, Vassilis ; Litsa, AkriviINSEAD Area: Technology and Operations ManagementIn: Journal of European Real Estate Research, vol. 1, no. 2, 2008 Language: EnglishDescription: p. 114-138.Type of document: INSEAD ArticleNote: Please ask us for this itemAbstract: Purpose The main scope of the paper is to demonstrate the capabilities of PYTHIA forecasting platform, to compare time series forecasting techniques, which were used to forecast mortgage loans in UK, and to show how PYTHIA can be useful for a bank. Design/methodology/approach The paper outlines the methods used to forecast the time series data, which are included in PYTHIA. Theta, the time-series used to forecast average mortgage loan prices, were grouped in: all buyers average loan prices in UK; first-time buyers average loan prices in UK; and home-movers average loan prices in UK. The case of all buyers average loan prices in UK, was presented in detail. Findings After the comparison of the methods, the best forecasts are produced by WINTERS and this is maybe due to the fact that there is seasonality in the data. The Theta method comes next in the row and generally produces good forecasts with small mean absolute percentage errors. In order to tell with grater certainty which method produces the most accurate forecasts we could compare the rest error statistics provided by PYTHIA too. Originality/value The paper presents the PYTHIA forecasting platform and shows how it can be used by the managers of a Bank to forecast mortgage loan values. PYTHIA can provide the forecasts required by practically all business situations demanding accurate predictions. It is designed and developed with the purpose of making the task of managerial forecasting straightforward, user-friendly and practical. It incorporates a lot of knowledge and experience in the field of forecasting, modeling and monitoring while fully utilizing new capabilities of computers and software| Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|
|
Europe Campus | Available | BC009011 |
Ask Qualtrics
Purpose The main scope of the paper is to demonstrate the capabilities of PYTHIA forecasting platform, to compare time series forecasting techniques, which were used to forecast mortgage loans in UK, and to show how PYTHIA can be useful for a bank.
Design/methodology/approach The paper outlines the methods used to forecast the time series data, which are included in PYTHIA. Theta, the time-series used to forecast average mortgage loan prices, were grouped in: all buyers average loan prices in UK; first-time buyers average loan prices in UK; and home-movers average loan prices in UK. The case of all buyers average loan prices in UK, was presented in detail.
Findings After the comparison of the methods, the best forecasts are produced by WINTERS and this is maybe due to the fact that there is seasonality in the data. The Theta method comes next in the row and generally produces good forecasts with small mean absolute percentage errors. In order to tell with grater certainty which method produces the most accurate forecasts we could compare the rest error statistics provided by PYTHIA too.
Originality/value The paper presents the PYTHIA forecasting platform and shows how it can be used by the managers of a Bank to forecast mortgage loan values. PYTHIA can provide the forecasts required by practically all business situations demanding accurate predictions. It is designed and developed with the purpose of making the task of managerial forecasting straightforward, user-friendly and practical. It incorporates a lot of knowledge and experience in the field of forecasting, modeling and monitoring while fully utilizing new capabilities of computers and software
Digitized
There are no comments for this item.