Modeling of Land Price at Kalaignar Karunanidhi Nagar in Chennai City

Abstract

This paper focuses on the modeling of land price using multiple regression technique and neural network technique. The model is used to forecast the unit residential land price of Kalaignar Karunanidhi Nagar (K.K.Nagar), which is located eight kilometers away from the core of Chennai City. The data sets are used to quantify the interaction behaviour and the share of contribution of influencing factors. The monthly average value of the selected factors such as National Gross Domestic Product, cost of crude oil, dollar equivalence to Indian currency, rate of inflation, gold and silver price, Mumbai and National share index, population in K.K.Nagar, interest rate on home loan, unit cost of construction, guideline value and time factor from the year 1997 to 2008 are considered to develop the models. Both multiple regression and neural network models are validated with the market price in the year 2009 and 2010. The results show that the average residual in the models is about 2 % higher and 2 % lesser respectively, which demonstrates the efficiency of the modeling. All factors contribute towards the model in neural network and out of thirteen, nine factors show good response in the multiple regression model. After validation the models are used to forecast the land price in the study area upto the year 2015. The Mean Absolute Percentage Error values of the multiple regression and neural network are about 0.43 % and 0.33 % respectively and the annual rise of land price in K.K.Nagar will be around 5 to 6 % in the next 5 years. Both the models are found to be well fit for the modeling of land price at K.K.Nagar, however the model using neural network shows better accuracy.