Using Thads Dataset to Determine the Impact of 2008 Subprime Crisis

 Summary Report: Difference in Market Value of Housing Units by Occupancy Status (THADS 2005–2013)

1. Introduction and Objectives

This assessment aimed to compare the Market Value (VALUE) of housing units based on their Occupancy Status (STATUS) (Occupied vs. Vacant) using the THADS data from 2005, 2007, 2009, 2011, and 2013. The objectives were twofold:

i.  To summarize the data using descriptive statistics and graphical methods.

ii. To test for a statistically significant difference in the mean market value between the two groups for each year.

2. Descriptive Statistics: Mean Comparison

Our initial descriptive analysis revealed a consistent pattern across all observed years:

  • Occupied units consistently exhibited a higher mean and median market value compared to vacant units. This suggests that, on average, properties that are occupied tend to hold greater market value. This could be attributed to various factors such as better maintenance, desirable locations, or the inherent value associated with an active household.
  • The market values for both occupied and vacant units showed considerable variability (high standard deviation and variance), indicating a wide range of housing unit values within each category. This implies that 'Occupied' and 'Vacant' categories encompass a diverse set of properties.

Impact of the Financial Crisis (2007-2011):

The data clearly shows the impact of the 2008 financial crisis on housing market values:

  • Pre-Crisis (2005-2007): Both occupied and vacant unit values generally increased, with occupied units maintaining a higher average value.
  • Example: Mean Occupied Value rose from ~$254k (2005) to ~$282k (2007).

During/Post-Crisis Downturn (2009-2011): A significant decline in market values for both categories was observed.

  • Example: Mean Occupied Value dropped from ~$282k (2007) to ~$245k (2009) and further to ~$233k (2011). Similar percentage drops were seen for vacant units.
  • Early Recovery (2013):  A slight recovery began to emerge, with mean values for both occupied and vacant units showing a modest increase from their 2011 lows.
  • Example: Mean Occupied Value slightly recovered to ~$240k (2013).

Throughout these fluctuations, the general trend of occupied units having higher market values persisted, though the absolute values changed significantly due to the market conditions.

3. Hypothesis Testing: Independent Samples T-Test

To formally test for statistically significant differences, independent samples t-tests were performed for each year, comparing the mean market value of occupied versus vacant housing units with a significance level (alpha) of 0.05.

2005: P-value = 0.1458 (Fail to reject H0) – No statistically significant difference.

2007: P-value = 0.2678 (Fail to reject H0) – No statistically significant difference.

2009: P-value = 0.7991 (Fail to reject H0) – No statistically significant difference.

2011: P-value = 0.8687 (Fail to reject H0) – No statistically significant difference.

2013: P-value = 0.0153 (Reject H0) – Statistically significant difference.

Key Finding from Hypothesis Testing:

Despite the consistent observation from descriptive statistics that occupied units tended to have higher market values, the t-tests revealed that this difference was only statistically significant in the year 2013. For all other years (2005, 2007, 2009, 2011), we failed to reject the null hypothesis, meaning that any observed differences in mean market value between occupied and vacant units could be attributed to random chance rather than a true population-level difference.

4. Conclusion and Impact on Housing Market

This assessment highlights an important distinction between observed differences and statistically significant differences. While occupied housing units generally had higher mean market values than vacant units across the entire period (2005-2013), this difference only became statistically significant by 2013. This finding suggests a potential shift in the housing market post-financial crisis:

  • Pre-2013 Stability/Uncertainty:  In the years leading up to and immediately following the 2008 financial crisis (2005-2011), the market may have been too volatile or the variations within occupied and vacant categories too large to establish a statistically robust difference in their mean market values, despite visible gaps.
  • 2013 – Emerging Clarity:  By 2013, as the market began its recovery and potentially stabilized, the distinction between occupied and vacant properties in terms of market value became clearer and statistically significant. This could reflect a renewed confidence in occupied properties, or perhaps a more pronounced devaluation of vacant or distressed properties as the market adjusted.

In summary, the 2008 financial crisis had a clear depressive effect on the market values of both occupied and vacant housing units. While occupied units generally maintained a higher average value throughout, it was not until 2013 that this difference was strong enough to be considered statistically significant, indicating a more definitive divergence in how the market valued these two categories of housing post-recession.

Comments

Popular posts from this blog

Teach your Child Coding

Google Data Analytics Project 2: Bike Share Case Study Project

ANALYZING CHANGE IN MEAN MARKET VALUE THADS DATASET