Scoring Model for Determining Investment Attractiveness as a Component of the Toolkit for Ensuring Investment Activity of Real Estate Funds

Abstract

Abstract. This article covers the analysis, scientific substantiation, and development of a scoring model for investment attractiveness of real estate assets which constitutes a comprehensive investment management toolkit of real estate funds. It has been established that existing approaches and practices for asset evaluation are predominantly focused either at exclusively financial metrics or qualitative expert judgment without adequate methodological formalization. Consequently, a complex multi-criteria scoring evaluation approach and a scoring model as such remain underexplored in both domestic and foreign academic literature. Based on conducted scholarly research and practical experience, a scoring model for investment attractiveness of real estate assets is proposed. The methodology consists of a multi-stage asset evaluation process, commencing with the structuring of a real estate longlist, verification of parameters’ categories, their weighting coefficients and evaluation metrics, and concluding with the conduct of expert assessment, calculation of a generalized integral score of investment attractiveness, and drafting the analytical conclusions. The model and its computation methodology incorporates six categories of key performance indicators – location, commercial, market, financial, physical condition of the asset, and the “other” category that have weighting coefficients assigned to each, and encompass the computation of 40 parameters in total. For the purposes of the obtained results and calcs interpretation, a five-level scale for investment attractiveness integral indicator of real estate assets is proposed. It was mapped based on the adjustments of the verbal-numerical Harrington’s scale; the critical threshold for asset selection is set at the level of 4.20 and above. An alternative implementation approach for the scoring model is proposed, which is based on a neural network, i.e. a multi-layer computational architecture that utilises artificial intelligence. The developed scoring model for investment attractiveness of real estate assets possesses the following advantages; it is concise, well-balanced, universal, relevant, and mirrors integrity. The scoring model should prospectively be examined from the empirical verification standpoint and focus on the development of automated implementation procedures.