Spatial Econometrics & Outlook

Stefan Jünger, Anne-Kathrin Stroppe, Dennis Abel

2025-04-24

Now

Day Time Title
April 09 10:00-11:30 Introduction
April 09 11:30-11:45 Coffee Break
April 09 11:45-13:00 Data Formats
April 09 13:00-14:00 Lunch Break
April 09 14:00-15:30 Mapping
April 09 15:30-15:45 Coffee Break
April 09 15:45-17:00 Spatial Wrangling
April 10 09:00-10:30 Spatial Wrangling
April 10 10:30-10:45 Coffee Break
April 10 10:45-12:00 Applied Spatial Linking
April 10 12:00-13:00 Lunch Break
April 10 13:00-14:30 Spatial Analysis
April 10 14:30-14:45 Coffee Break
April 10 14:45-16:00 Spatial Econometrics & Outlook

What are spatial econometrics?

Classic econometrics:

  • Using statistics to model (complex) theories, esp. causal thinking
  • As default, we think about regression analysis

One core assumption: Observations are independent of each other. However, we just learnt that is often not the case.

   

Where does spatial dependence and spatial processes enter our models and affect our outcome of interest?

Is it meaningful or just nuisances?

    Space can be important in our analysis in two ways.

    • It’s meaningful in our theory, and we thus interpret it accordingly after estimation
    • It can distort our empirical estimates, producing bias, inconsistency, and inefficiency

    We can address these different perspectives in our analysis with spatial econometric methods.

Spatial Diffusion

  • \(y_i\) affects \(y_j\) through \(w_{ij}\)
  • \(y_j\) affects \(y_i\) through \(w_{ji}\)
  • endogenous by design!
  • Examples:
    • tax competition: if a state cuts corporate tax, neighbours respond by cutting theirs too
    • civil war onset: conflict in one country raises the probability of onset in neighbours

Spatial Spill-Over

  • \(x_i\) affects \(y_j\) through \(w_{ij}\)
  • \(x_j\) affects \(y_i\) through \(w_{ij}\)
  • Examples:
    • trade and export: a neighbour region’s GDP (their X) raises your export volumes or wages (your Y)
    • crime displacement: increased policing in one neighbourhood raises crime in your neighbourhood, as criminals relocate

Formulas…

    Linear Regression: \[\small Y = X\beta + \epsilon\]

    Spatial Lag Y / Spatial Autoregressive Model (SAR, Diffusion): \[\small Y = \rho WY + X\beta + \epsilon\]

    Spatial Lag X Model (SLX, Spillover): \[\small Y = X\beta + WX\theta + \epsilon\]

    Spatial Error Model (SEM): \[\small Y = X\beta + u\] \[\small u = \lambda Wu + \epsilon\]

Flavors and extensions

But what if….

  • … you have interdependence and spillovers in covariates?
  • … spillovers and clustering in errors?
  • … interdependence and clustering in errors?
  • … everything is related with everything?