Market Validation of Automated Value Models for Real Estate in Scotland

ACHMAD TRI BUDIAWAN, Roger Adam; Najib Murad; Bowo Setiyono, S.E., M.Com., Ph.D.

2023 | Tesis | Magister Manajemen

Industri properti telah tumbuh secara signifikan.  Akurasi nilai berpengaruh pada permintaan dan penawaran sebagai faktor kunci dalam pasar properti.  Mereka yang memprediksi harga lebih akurat dapat menghasilkan lebih banyak keuntungan.  Akurasi harga diperlukan sebelum membuat kesepakatan untuk penjual, pembeli, penyewa, tuan tanah, dan hipotek.  Perusahaan dan platform penyedia informasi harga properti; beberapa di antaranya memanfaatkan disrupsi teknologi dengan memberikan informasi kepada pengguna secara online, seperti Rightmove, Zoopla, dll.  Namun, perkiraan harga tersebut disebut masih jauh dari akurat.  Hal ini disebabkan oleh sumber daya, pendekatan, dan metodologi dalam memprediksi harga properti.   Penelitian ini membantu Property Price Hub Ltd (PPH) memperoleh keunggulan kompetitif yang berfokus pada penyediaan informasi harga properti yang lebih akurat, cepat dan mudah kepada pelanggannya.  Peneliti berusaha memberikan anternatif terhadap metode penilaian properti yang telah umum digunakan seperti pendekatan biaya, pendekatan pendapatan, dan pendekatan data pasar properti di Skotlandia.  Memanfaatkan ketersediaan data transaksi properti yang terekam rapi, legal, serta mudah diakses, peneliti mencoba membuat Automated Value Model (AVM) memanfaatkan machine learning untuk memprediksi harga properti.  Peneliti menggunakan software Phyton 3.0 (pycaret lib) serta  Orange untuk mengolah Price Paid Data (PPD) yang disediakan oleh HM Land Registry untuk membuat model prediksi.

The real estate industry has grown significantly. Even in the midst of a pandemic, it could survive inflation. Accuracy of market value as the foundation for demand and supply creation is one of the key factors in the real estate market. Those in business who can accurately predict pricing can make more money than those who estimate prices incorrectly. Price accuracy is necessary before making a deal for sellers, buyers, tenants, landlords, and mortgages. A company that gives information on the market price of a property is thus currently emerging as a new business sector in the real estate market. Several companies and platforms are currently recognized that provide information on property prices; some of them take advantage of technological disruption by giving users information online, such as Rightmove, Zoopla, etc. However, the current price estimates from information providers are still far from accurate. This can be caused by the resources, approach, and methodology used in predicting property prices. Property Price Hub Ltd, a new private venture based in Edinburgh is trying to provide a solution to this problem. Armed with machine learning-based data processing capabilities, utilizing several API features and geovation networks, PPH has a great opportunity to enter the property market to compete with existing businesses. However, PPH still requires a review of the methodology to be used as well as market validation regarding quality products and services that are attractive to potential customers.

Kata Kunci : market value, machine learning, property prices