Thesis (M.U.D.)--University of Hong Kong, 1991. / Includes bibliographical references (leaves 153-161) Also available in print.

Thesis (M. Hous. M.)--University of Hong Kong, 2006.

No description available.

The effect of parks on residential real estate has been well recorded; however little research has been done to estimate the effects of parks on commercial real estate values. With the help of Geographic Information Systems and the transactions of nearly 10,000 properties from 2000 to early 2011, I produced three different hedonic commercial real estate sales models. Controlling for building characteristics, demographic variations within census groups, and locational attributes, I find that proximity to parks plays a role in the valuation of some property types. Little evidence is found to support the hypothesis that properties in proximity to parks are valued higher; however co-location of parks and other attributes could play a beneficial role in supplying cities with more parks while boosting surrounding property values.

In a unique opportunity to examine rare appraisal data from the commercial real estate sector, the accuracy of three machine learning algorithms is compared in the task of appraising commercial real estate. The algorithms; random forests, support vector regression and artificial neural networks, are tested in research about residential real estate, but the area of commercial real estate has remained relatively unexplored due to corporate secrecy. The mean absolute percentage error of the trained models range from 44% to 24% and is held as a baseline. The best performing baseline model, Random forests, was then made more sophisticated in order to evaluate how much performance could increase. It was found that the introduction of Gradient boosting reduced the aforementioned error from 24% to 20%. In comparison, the average human expert appraiser performs at an average error of 12%. The conclusion is that more work is needed in order to compete with human expert appraisers - and that this is a feasible task considering some of the inherent issues within the used data could be resolved with much manual labor. / Med hjälp av en unik datamängd från den kommersiella fastighetssektorn utvärderas prestandan hos tre maskininlärningsalgoritmer i uppgiften att värdera kommersiella fastigheter. Dessa algoritmer; slumpmässig skog, stödvektormaskin samt artificiella neurala nätverk, förekommer ofta inom forskning vid värdering av privata boenden, men på grund av datasekretess inom företagsvärlden är det kommersiella fastighetsområdet idag relativt outforskat. Den genomsnittliga procentuella felmarginalen för de tränade modellerna ligger inom intervallet 44% till 24% och detta hålls som en baslinje för prestandan. Den bäst presterande modellen, slumpmässig skog, görs sedan mer sofistikerad i syfte att utforska hur mycket prestandan kan öka. Det konstateras att en implementering av så kallat Gradient boosting sänker den tidigare nämnda felmarginalen från 24% till 20%. Jämförelsevis presterar den genomsnittliga mänskliga värderaren vid en felmarginal omkring 12%. Slutsatsen är att mer arbete krävs för att konkurrera med mänskliga värderare - och att detta bedöms vara genomförbart med tanke på att vissa underliggande problem i den använda datamängden kan lösas med en stor mängd manuellt arbete.

published_or_final_version / Real Estate and Construction / Doctoral / Doctor of Philosophy

Abstract This research paper tracks the growth of e-commerce retail sales from 2000 through 2018 to see if the emergence of e-commerce has affected the retail real estate market in any way. With many stores now offering online shopping and many large retailers contemplating lowering their presence in brick and mortar stores, this question is relevant now more than ever. In order to test for the correlation between e-commerce retail sales growth and the retail real estate market, a collection of retail real estate factors are investigated. This paper tests retail real estate factors at the United States country level and a more in-depth city level that investigates 66 individual cities nationwide. In the OLS modeling, these retail variables represent my dependent variables. These dependent variables were then regressed on a yearly basis against total e-commerce retail sales and a vector of control variables. The results show that returns of the NCREIF property index, asking rent, and percent change in commercial real estate prices for the U.S. are negatively impacted by total e-commerce retail sales. These results indicate that the growth of e-commerce retail sales has negatively impacted the retail real estate market to some degree. This correlation can have long term impacts on the survival of many retail businesses and can eventually affect the economies of small towns and cities across the country. If physical retail locations begin to suffer, other local stores and restaurants will likely suffer as well, eventually impacting the economy on a larger scale.

Thesis (M.U.D.)--University of Hong Kong, 1992. / Includes bibliographical references.

published_or_final_version / Housing Management / Master / Master of Housing Management

published_or_final_version / abstract / Real Estate and Construction / Doctoral / Doctor of Philosophy