Analysis of house prices in Chicago

Analysis of house prices in Chicago

Introduction

The real estate industry is a volatile market whereby housing prices can change anytime. Tupenaite et al. (2017) report that housing prices throughout the world have not been consistent since when countries like Germany and Japan have not registered housing booms, countries like France, Spain, and the UK have experienced an increase in housing prices over the same period. Glaesar & Gyourko (2018) assert that positive economic growth leads to a property boom in the USA. Some of the factors that influence the real estate sector in the country are the Consumer Price Index (CPI), per capita income, and mortgage rate.  House prices being the dependent variable, we seek to understand how independent variables; prime rate, inflation rate, per capita income, mortgage rate, number of Starbucks openings, Dow Jones Industrial Average (DJI), and change in mortgage rate have contributed to the rise of house prices in Chicago downtown. This makes up my hypotheses.

Literature review

Economic growth has influenced housing prices in America. According to Gurran et al. (2015), houses, unlike other commodities, are not affected by the forces of demand and supply since locations are the influencers of housing prices. Falcon (2019) claims that their proximity to wholefood stores influences housing prices. This is backed by the fact that investors who have properties near the stores recorded an increase in return on investment over five years. Ivanova (2018) agrees with Falcons (2019) and affirms that places, whereby Starbucks has set shop, have recorded an increase in house prices. Ivanaova (2018) reports that in high-end cities like New York and San Francisco, the presence of coffee shops leads to a rise in housing properties by $6500. Capps (2014) reports that anywhere that houses within a half-mile radius of where Walmart has set shop have recorded a 2-3% increase in house prices. An assertion by Özmen, Kalafatcılar, & Yılmaz, (2019), house prices have been on the comparative rise with increased per capita income, population growth, which brings about high demand and overall wealth of people. However, high mortgage rates have always had a negative relationship with the prices of houses. Ashish (2016) claims, having quantifiable predictors, it is easy to explain the price variations in real estate products using hedonic regression analysis. In this paper, I will utilize the regression technique to examine the relationship with independent variables.

Data and the sample

The data set for this analysis was obtained over the internet on an open data source that contained the monthly record of data from the year 2016 to 2018. I found this data fundamental in explaining the current state house prices since it is the most recent. This dataset can also be used as time-series data, which is very important in comparing the relationship between house prices with time. The variables contained in this data set have been stated above, which will be analyzed to find answers for the hypothesis.

Hedonic regression equation.

This regression technique makes use of statistical regression to measure one dependent economic variable with one or more independent economic variables. In this context, the response variable is the house prices recorded for thirty-six months from January of 2016 to December of 2018. The function takes an ordinary statistical multiple regression equation of the form  Where Y is the response variable,  is the intercept, and  are the coefficient values of the predictor variables  that explains the variations in the response variable. The adequacy of the estimated regression line depends on the value of the coefficient of multiple determination. The negative and positive signs of a coefficient explain the amount of decrease or increase of an independent variable that leads to one unit increase of a dependent variable.

Descriptive analysis.

Descriptive statistics are simply the statistics that inform about the concentration and dispersion of the variables data points. The descriptive analysis includes the measure of central tendencies like mean and the median and measure of distribution, such as the variance and the standard deviation. The mean can inform us on the average house prices, per capita income, consumer price index, and other variables for the recorded months as well their variances and standard deviation. The standard deviation tells how the amounts in a given variable are scattered from its mean. For example, the standard deviation of house prices scatters up to 14434.364 from the mean 403166.67. The table below outputs the descriptive statistics of our data set.

Summary statistics:

Correlation analysis and visualization of scatter plots.

Correlation analysis is a crucial statistical technique that helps in investigating the degree at which two variables move relating to one other. There are three kinds of correlation coefficients, namely, Kendall, Pearson, and spearman. The values of all the three ranges between -1 and +1, which means if the correlation coefficient between variables in negative, then the two moves in inverse direction with one another, whereas if the value of positive, then the two variables move in direct proportional direction in relation to one another (Benamraoui, 2018). The table below shows the correlation between house prices and the independent variables.

Correlation matrix:

Change in the mortgage rate is the only predictor variable that has no significant correlation with the house price. The other variables, especially the inflation rate and per capita income, have indicated a powerfully positive relationship with the dependent variable.  To have more insights on the correlation value above, it was wise to visualize the relationships of the dependent and independent variables using the scatter plots below.

From the scatter plots, we can visualize strong relationships between house prices and inflation rate, per capita income, and DJI. The number of Starbucks opening variable looks like an indicator variable and has no much contribution. Prime rate and change in mortgage rate have no clear linear relationship with house price variable as it can be seen from their respective plots.

Time series and analysis and moving average graphs

Time series analysis is an essential statistical tool used to examine time-series data. We can regard our data set as time series due to the existence of a date column that record time in months from 2016 to 2019. This analysis is essential because, based on the time, we can be able to predict house prices beyond the recorded time. The table below shows the moving average graphs for increased house prices.

We can visualize that there has been an increase in the costs of houses with time. This can also be associated with the rise in the consumer price index and per capita income with time. This is shown with their respective moving graphs below.

Regression analysis

As pointed above in the hedonic regression equation section, regression tries to show the relationship between two or more variables and how an increase or decrease of one or more inputs affects a specific output. The regression line is fitted by the least square method, which works to reduce mean residual errors of data points, and the accuracy is measured by the R-squared value of the model (Owusu-Ansah, Adolwine, & Yeboah, 2017). The following two regression lines were fitted to find a better fit.

The R-squared value 0.9799 shows that 97.99% of house prices in Chicago are due to movements or variations of the independent variable used in the model. The equation shows what amount increases or decreases by a predictor variable results in 1 unit increase the dependent variable when other variables are constant. For example, a 1 unit increase in house prices is caused by 7.0155 increases in per capita income when other variables are kept constant. A positive sign indicates an increase, while a negative sign of coefficient indicates a decrease in a predictor variable.

The R-squared value 0.9693 means that 96.93% of house prices are due to variation of DJI, inflation rate (CPI), and per capita income. The regression line says 1 unit increase in house price is attributed by the increase or decrease of a predictor amount represented by its coefficient. For example, a 1 unit increase in price is due to a 0.7212 decrease in DJI.

The first model included all the predictor variables except change in mortgage rate due to its weak correlation with house prices, while the second model comprises only three predictors that showed high relationships with house prices. Since there is no much difference in the R-squared between the two models, both can be used to measure real estate prices.

Findings

From the analysis, it was evident that the prices of houses moved hand in hand with the state of the economy in Chicago. For the last three years, there has been a significant increase in consumer prices index and per capita income, which characterized the increased prices in the real estate product. According to the literature reviewed, closeness to wholefood stores and locations that had easy access to essential services were characterized by high house prices, which shows that big chain companies have contributed to the rising costs of real estate products.

Conclusion

House prices in Chicago have witnessed a continuous increase in a factor that directly affects individual people, businesses, and institutions that consumes real estate products. This study purposed to investigate how economic factors such as consumer price index, per capita income, mortgage rates, prime rate, and DJI affects the pricing of houses in Chicago. Descriptive statistics were produced to help understand the distribution of house price variables and the independent variables. To understand the linear relationships, correlation analysis was conducted between the dependent variable and each independent variable and visualized using scatter plots. Time series and regression analyses were done to uncover the relationship between housing prices with time and the effects of economic factors on real estate prices. The first regression line is a better estimator of prices with an accuracy of approximately 98%. As a result, the hypothesis answer is that the used variables have a significant influence on house prices.

References

Bourassa, S. C., & Hoesli, M. (2017). HIGH-FREQUENCY HOUSE PRICE INDEXES WITH SCARCE DATA. Journal of Real Estate Literature, 25(1), 207-220. Retrieved from https://search.proquest.com/docview/1926546984?

Benamraoui, A. (2018). A comparative study between the UK and the USA house price indicators before and during the financial crisis of 2007-2009. Journal of Financial Economic Policy, 10(4), 456-472. doi:http://dx.doi.org/10.1108/JFEP-02-2018-0025

Capps, K. (2014). Want to Boost Housing Values? Build a Walmart. Retrieved 19 November 2019, from https://www.citylab.com/equity/2014/11/want-to-boost-housing-values-build-a-walmart/382381/

Falcon, J., Howley, K., Howley, K., Smith, M., Lane, B., & Falcon, J. et al. (2019). This is how grocery chains affect a home’s value – HousingWire. Retrieved 19 November 2019, from https://www.housingwire.com/articles/49767-this-is-how-grocery-chains-affect-a-homes-value/

Glaeser, E., & Gyourko, J. (2018). The Economic Implications of Housing Supply. Journal Of Economic Perspectives, 32(1), 3-30. doi: 10.1257/jep.32.1.3

Gurran, N., Phibbs, P., Yates, J., Gilbert, C., Whitehead, C., Norris, M., … & Rowley, S. (2015). Housing markets, economic productivity, and risk: international evidence and policy implications for Australia—Volume 2: Supplementary papers. Australian Housing and Urban Research Institute (AHURI), 255, 1-116.

Ivanova, I. (2018). This is how much Starbucks adds to the price of a nearby home. Retrieved 19 November 2019, from https://www.cbsnews.com/news/starbucks-makes-property-values-jump-study-shows/

Owusu-Ansah, A., Adolwine, W. M., & Yeboah, E. (2017). Construction of real estate price indices for developing housing markets: Does temporal aggregation matter? International Journal of Housing Markets and Analysis, 10(3), 371-383. doi:http://dx.doi.org/10.1108/IJHMA-06-2016-0047

Özmen, M., Kalafatcılar, M., & Yılmaz, E. (2019). The impact of income distribution on house prices. Central Bank Review, 19(2), 45-58. doi: 10.1016/j.cbrev.2019.05.001

Tupenaite, L., Kanapeckiene, L., & Naimaviciene, J. (2017). Determinants of Housing Market Fluctuations: Case Study of Lithuania. Procedia Engineering, 172, 1169-1175. doi: 10.1016/j.proeng.2017.02.136

Appendix

Chicago real estate and economic dataset