May 25, 2016

Project Construct

Can gentrification be predicted?

  • Test Location: Essex County, NJ, municipality level (n = 22)
  • Data Source 1: Residential property sales prices (NJ Dept. of the Treasury)
  • Data Source 2: Mean household income (ACS, US Census Bureau)

Hypothesis

In areas that are on the verge of gentrifying, the mean home sales prices will increase at a much higher rate than mean household incomes. This could mean that investors are starting to speculate in that area, purchasing homes at increasingly higher prices, while the residents of the area have not yet changed (so income levels are more static, or growing much more slowly).

Essex County, NJ

Time Series Plots I

Time Series Plots II

Time Series Plots III

Ratio Plots I

Ratio Plots II

Ratio Plots III

Forecasting (Exponential Smoothing)

The simplest form of exponential smoothing is given by the formula:

\[s_{t} = \alpha * x_{t} + (1-\alpha) * s_{t-1}\]

where \(\alpha\) is the smoothing factor, and \(0 < \alpha < 1\). In other words, the smoothed statistic \(s_{t}\) is a simple weighted average of the current observation \(x_{t}\) and the previous smoothed statistic \(s_{t−1}\).

Sample Forecast (West Orange)

##      Point Forecast    Lo 80    Hi 80    Lo 95    Hi 95
## 2015       198.4573 184.3748 212.5399 176.9200 219.9947
## 2016       198.4573 179.8160 217.0986 169.9479 226.9667
## 2017       198.4573 176.1634 220.7513 164.3616 232.5530
## 2018       198.4573 173.0233 223.8913 159.5594 237.3553
## 2019       198.4573 170.2244 226.6903 155.2787 241.6359
## 2020       198.4573 167.6733 229.2414 151.3772 245.5375
## 2021       198.4573 165.3128 231.6019 147.7671 249.1475
## 2022       198.4573 163.1047 233.8100 144.3901 252.5246
## 2023       198.4573 161.0219 235.8927 141.2048 255.7098
## 2024       198.4573 159.0447 237.8699 138.1809 258.7337

Forecast Plots

Motion Chart

Regression Coefficients I

Municipality Coefficient Municipality Coefficient
Belleville -2.96 Irvington 5.86
Bloomfield -1.15 Livingston -1.41
Caldwell -0.18 Maplewood 1.42
Cedar Grove 2.59 Millburn -2.39
East Orange -7.59 Montclair 0.15
Essex Fells -1.31 Newark -2.61
Fairfield -1.63 North Caldwell 0.19
Glen Ridge 0.52 Nutley -0.66

Regression Coefficients II

Municipality Coefficient
Orange 1.04
Roseland 0.43
South Orange -2.06
Verona -0.31
West Caldwell 1.49
West Orange 4.78

Conclusions

Can gentrification be predicted?

  • Higher regression coefficients seem to point to more movement
  • Forecasts of more stable ratios are narrower
  • As communities gentrify, income and home sale prices should reach an equilibrium point
  • Communities near the median have ratios somewhere between 0.5 and 0.7; ratios that fall outside that range may indicate where economic change is happening at a faster rate than average
  • Economic downturn skews trends; need additional pre- and post- data