Changes in the Real Estate Market - Case Study Example

Understanding the Real Estate Market: Determinants of House Prices Author: Course Title: MBA 5103, Business Statistics Program: PMBA Name of your Institution: HCT [Instructor Name] Date: 26/10/2014 Executive Summary The value of house depends on the characteristics of the house. Each characteristic has different values that contribute to the value of a house. A house becomes more valuable when these variables are integrated. The results show that all variables were highly significant and related to the price of the house. Table of Contents Executive Summary 2 Introduction and background 4 Methodology 6 Analysis and discuss of results 7 Normal distribution 7 Normal distribution of house prices 8 Regression of house and interpret the R2 of the regression 8 The estimated housing equation 9 Use of model to develop strategies for growth and expansion 11 Conclusion 11 References 12 Appendix 13 Introduction and background There have been changes in the price of houses for the last decade. There has been a greater impact on demand and supply of houses. Buyers are ever demanding houses that meet their demands while house owners continuously search for sustainable returns from their products. The nature of demand for houses is significantly changing because of both external economic shifts and the overall organization change response. These have greatly affected the way houses have been packaged and supplied. There has been much debate concerning about disconnect between needs of buyers and the house prices. Property owners have been blamed for not adjusting to clients demand and understanding their increasing need for flexibility and diversity. Several factors, not related to the macro-economic conditions, also affect price of houses and include number of bedrooms, land area of the house, existence of a pool, existence of garage and number of Bathrooms. The number of bedrooms and land area of the house are important factors to consider when determining price of a house in the real estate market. There have been micro-economic articles that have focused on the models of the value of houses and impact of those factors where most have focused on property characteristic issues that affect prices. House price is an important aspect of property market that should be researched. It is worth pointing out that house prices is sensitive to space area and number of bedrooms requirements assumptions, availability of garage, pool and bathroom. House price analysis has attracted interests from buyers, lenders, appraisers, urban planners and other stakeholders with the main aim of understanding factors that influence real estate market. House market experienced uncertainty volatility in prices, which leads to financial losses. Many studies have been conducted on house market making it the most researched topic in the real estate sector and in is attributed to the size, importance and the availability of data. There is still a gap in actually finding how price and house characteristic interrelate in market. Many microeconomic articles that affect have also discussed the impact of special issues on house prices. Leishman (2000) did an examination of the structure of house prices. They came up with conclusions on how prices vary depending on a property distance from the city centre. According to Sayer and Moohan (2007) carried out a study of the viability of downtown properties in the context of growing city communities, the study established that properties has an important role in local economies and that the demise of city centre properties in not imminent. Hough and Kratz (1983) examined physical characteristics of the properties. Hetherington, (2011) examined architecture features and how their differences affect prices. This study also falls under microeconomic literature of office-property research. This paper will add understanding to the factors that influence price and number of houses on the market. Land size and number of bedrooms are highly correlated since taller buildings are larger. The correlation between size and number of bedrooms, existence of pool, garage and number of bathrooms in the current studies is about 85%, which has been consisted based on earlier predictions. Methodology A mathematical model of regression can be proposed that would reflect the dependence of price on number of bedrooms, land area of the house, existence of a pool, existence of garage and number of Bathrooms. The formal model can be developed as below: Y = β0+ β1B+ β2S + β3P + β4G + β5T+ ui In the above regression equation Y is the price of a house is in 1000’s of dollars, ‘β0’ is the constant/intercept term, B means number of bedrooms, S means land area of the house, P means existence of a pool, G means existence of garage and T means number of Bathrooms, β1, β2 β3 β4 and β5 are coefficient of the variable and ui is the error term. The error term is included in the regression equation because practically all of the variation in the dependent variable can not be explained by independent. In a regression model it is not possible to incorporate all possible factors that affect the dependent variable. The error term, is incorporated in a regression model in order to capture the effect of those factors on the dependent variables that are not included as the independent variable in the regression equation (Berk, 2003) The basic assumptions made for running the above regression are as follows: a. Mean of the disturbance or the error term would be zero, i.e. E(ui) = 0. b. Variance of the error term is constant for all i, i.e V(ui)=σi, for all i. c. ui and uj are completely independent of each other. The regression model developed above is a multiple regression model as it incorporates more than one independent variable. In this equation β1 represents the rate at which price changes following a change in number of bedrooms, considering other factors as constant. In other words to say β1 measures the effect of number of bedrooms on price after eliminating the effect of other independent. The same applies to other coefficients. In the analysis price of a house is in 1000’s of dollars so that a price of 230 means the price of the house is $230,000. If a house has a pool it takes a value of 1 if no pool it takes a value of 0. Similarly, with respect to garage, 1 means there is garage and 0 means no garage. Size refers to the land area of the house including non-built area in square feet. Analysis and discuss of results Summary of descriptive statistics for house prices House prices 1 count 105 2 mean 221.1029 3 sample standard deviation 47.1054 4 sample variance 2218.919 5 minimum 125 6 maximum 345.3 7 range 220.3 8 standard error of the mean 4.597 9 confidence interval 95.% lower 212.0929 10 confidence interval 95.% upper 230.113 11 skewness 0.474 12 coefficient of variation (CV) 21.30% 13 1st quartile 187 14 median 213.600 15 3rd quartile 251.4 16 interquartile range 64.4 Normal distribution For a normal distribution, the empirical rule says that 95.44% of the observations are within the range of ±2*σ +μ. Calculate the value ±2*s +μ for house prices. How does your result compare with the empirical rule? ±2*s +μ = ±247.1054 +221.1029 = 315.3137 Between 126.8921 and 315.3137 The results is not different from the expectations Normal distribution of house prices Furthermore, by observing the values of the variable collected over the period it could be concluded that the data is normally distributed as majority of the values lie within two standard deviations of the mean value. This is further tested by comparing the mean, median and mode values obtained this show that mean, and median and are very close to each other which is supportive of the normal distribution of data. Regression of house and interpret the R2 of the regression Regression Analysis R² 0.598 Adjusted R² 0.577 n 100 R 0.773 k 5 Std. Error 30.974 Dep. Var. Price ANOVA table Source SS df MS F p-value Regression 134,145.7513 5 26,829.1503 27.97 2.98E-17 Residual 90,181.9962 94 959.3829 Total 224,327.7475 99       Regression output confidence interval variables coefficients std. error t (df=94) p-value 95% lower 95% upper Intercept 10.8578 33.5975 0.323 .7473 -55.8509 77.5664 Bedrooms 7.5451 2.4127 3.127 .0023 2.7546 12.3356 Size 0.0358 0.0143 2.504 .0140 0.0074 0.0642 Pool 20.3605 6.5955 3.087 .0027 7.2650 33.4561 Garage 42.0254 6.8757 6.112 2.22E-08 28.3735 55.6774 Baths 30.2961 8.6345 3.509 .0007 13.1521 47.4400 The stepwise regression equation above have R2 (coefficient of determination) of 0.598 which suggests that the stepwise regression performed between variables is only able to predict 59.8% of the total variations observed in the values. Thus 59.8% of the variability is explained by the linear model. It accounts a moderate fit of the model to the data. This could be due to the certain values that could be considered as outliers observed within the data over the period of analysis. The estimated housing equation Y = β0+ β1B+ β2S + β3P + β4G + β5T+ ui Price = 10.85578+7.5B+ 0.0358S + 20.3605P +42.0254G +30.2961T Whereby B means number of bedrooms, S means land area of the house, P means existence of a pool, G means existence of garage and T means number of Bathrooms. i. Using α = 0.05 and the appropriate statistic, test the hypothesis that b1≠ 0; b2≠ 0 ; b3≠ 0; b4≠ 0; b5≠ 0, (test whether coefficients are individually significant) . Reject H0 if tcalc < t crit for any variable Do not reject H0 if tcalc > t crit for any variable Where tcalc is t calculated and t crit is t critical Critical value at 0.05 or 95% level of significance Reject H0 if tcalc > t crit (1.96) Do not reject H0 if tcalc < t crit (1.96) Looking at the t values of the variables it can noted that bedrooms have 3.127, land area of the house has 2.504, existence of a pool has 3.087, existence of garage 6.112 and number of Bathrooms has 3.509. Since t calc > t crit (< 1.96), accept reject H0. Since all the coefficients of the variables have their t-values greater than the 1.96. Then we accept the hypothesis and conclude that coefficients are individually significant in determining the price of the house. Positive coefficients mean that the more an increase in value in independent variable, the higher the price. ii. Using α = 0.01 and the appropriate statistic test the hypothesis that b1 = b2 = b3 = b4= b5= 0 (test whether coefficients are jointly significant) Looking at overall regression model it will be noted that p-value of the F-test is 2.9810-17 which is lower than 0.01 and the F-test value is 27.97 which is higher than the critical value of 2.7. Hence the coefficients are jointly significant for population data. Use of model to develop strategies for growth and expansion The model can be used in predicting the price of houses they have before marketing them. They will estimate the house prices using this model and avoid scenarios where they will stay with property for a long period because of high prices Conclusion The aim of analysis was to understand whether a relationship exists between price on number of bedrooms, land area of the house, existence of a pool, existence of garage and number of Bathrooms. The findings show that number of bedrooms, land area of the house, existence of a pool, existence of garage and number of Bathrooms influence house price. Overall, the empirical model developed from a brief review of earlier literature helps explain trends in prices. The descriptive statistics made life easier in terms of showing effects of model. With all the information given from the tables, we ended up with these observation house prices do not change a lot when the when one unit of land size is introduced. This can clearly be proved by comparing the standard deviation, standard error and range. Regarding the Confidence Intervals, a lot can be inferred from the data provided. It can be seen that confidence interval for is not a good indicator to be used as a general tool for analysis. Lastly, the results do confirm our suggestion that we put forward in the beginning of the experiment. Many of the tests performed added value of the findings that were obtained, and some of the hypothesis were actually rejected while others were accepted. It was found that the confidence interval for the data lies similar to the findings of the linear regression. This proves that the graphical method sometimes is the best was to figure out something about a pattern. References Berk, R. A. (2003). Regression Analysis: A Constructive Critique. London : Sage Guo, S. & Fraser, M. (2014). Propensity Score Analysis: Statistical Methods and Applications. New York: SAGE Publications Hetherington, A. , (2011). Central London Property Market Review. CBRE Research | Quarter 4 201. Available online at [Accessed 25 October 2014]. Hough, D. & Kratz, C.,1983. Can ‘Good’ Architecture Meet the Market Test?, Journal of Urban Economics, 1983, 14, 40–54. Leishman, C., 2000. Estimating Local Housing Market Price Indices Using Land Registry Data. Property Market Economics Group, Department of Building, Engineering and Surveying, Heriot-Watt University. Housing Finance No. 47 Sayer, J. & Moohan, J., 2007. An Analysis and Evaluation Of Hedonic Price Valuations In Local Leasehold Office Markets. Available online at [Accessed October 2014]. Wolldridge, J. (2012). Introductory Econometrics: A Modern Approach. New York: Cengage Learning. Appendix In excel sheets Read More