Data Science and Machine Learning for Real Estate class

Data Science and Machine Learning for Real Estate, a nine-week, spring semester elective course offered by the MIT Center for Real Estate, produced the second group of data scientists joining the real estate industry workforce. Students learned the basics of data science, econometrics, and predictive analytics, how to code in R, and how to apply knowledge and skill to solving complex questions in real estate. The class featured industry leaders from Jones Lang Lasalle (JLL), Compstack, StateBook, and Cherre, who helped students learn how data leads to better decision making across the industry. As a culmination of their course work, students explored a topic of their choice and showcased them in a poster format.

What are the different ways data science can help the real estate industry? As the only data science course offered within the MIT Center for Real Estate, this elective provided students an opportunity to embrace, appreciate, and utilize wide data across numerous aspects of the real estate industry. Over the course of nine weeks, under the guidance of Dr. Andrea Chegut and teaching assistants Jim Peraino and James Scott, students learned to code in R and to listen to every data point as a story of its own.  From the data cleaning process to the use of econometrics in answering real estate questions, the course introduced a full process of data science so that students understand how to conduct their own data science project upon the course completion. To help this effort, the course provided eight R notebooks to be used as guidebooks, which entailed various real estate data sets and a data visualization toolkit.

Representatives from various leaders in the real estate data science and technology domain – namely, Jones Lang Lasalle (JLL), Compstack, StateBook, and Cherre – served as guest speakers in the class.  These presentations helped students understand how data leads to better decision making across the industry. To conclude the semester, students utilized Compstack data to demonstrate their ability to identify patterns in a conjoined data set and to use predictive analytics.  Their findings were showcased in a poster format and presented as their final project.