Search

location'sdemandandthepurchasers'affordabilitydeterminetheproperty'soptimalvalue.To meet this optimal price the facilities in the properties are arranged through house sorting. Before sorting, it is important to find the significant facilities that contribute to the property value. The previous works mostly through qualitative assessment have identified the important factors influencing residential property value. The current article aims to determine the significant property facilities that are contributing to its value using statistical techniques. The facilities of a residential property considered in this study are Total Area, Age, Maintenance Cost, Number of bedrooms, Number of Toilet, Number of Balconies, facing of the property, Furnishing, Availability of Power Backup, Security, Elevator, Parking facilities. Data needed for the study were collected through a ground survey and interviews with the owners and real estate agents in the Greater Hyderabad Municipal Corporation (GHMC) Region. A total of 202 residential apartments that are yet to be sold were considered for the model development and 32 sold-out properties were obtained to validate the models. The significant facilities are identified using the p-test and the Group Methods of Data Handling- Neural Network (GMDH-NN) and linear regression techniques are adopted to develop the models for all possible combinations of the most significant factors. The accuracy of the developed models is assessed using R2 and Root Mean Square Error (RMSE) and Mean Average Percentage Error (MAPE). The study's findings indicate that the total area of the Residential apartment and the maintenance cost are the most significant factors. Both GMDH-NN and linear regression models developed using these factors show that the consideration of the Total area alone as a variable for prediction of property value has R2 > 0.748 and the RMSE and MAPE being less than 30%. The accuracy of prediction for the validation datasets using the Total area of the property as an input variable is more than 70% for both models. Identifying the significant factors using statistical techniques and then developing the models using these factors is the novelty of the research. the proposed models in the article can be used directly to understand the preferred facilities in Residential properties in the GHMC region and the adopted framework can be used for property valuation in general for any region