Dataset

Schools

Student-teacher longitudinal data with anonymized identifiers and synthetic regressors/outcome.

All datasets
Bipartite incidence pattern for Schools

Graph Summary

Two-way graph uses student and teacher identifiers.

The primary benchmark graph is built from unique (id1, id2) pairs. The figure shows a binned sparsity pattern of the bipartite incidence block, with both partitions relabeled to contiguous integer identifiers. This block is the off-diagonal part of the corresponding graph Laplacian.

413,444rows
406,998unique edges
206,722id1 levels
11,960id2 levels
15components
218,682nodes
CSV ZIP Stata DTA

Variables

Columns in the clean CSV:

id1 id2 id3 x1 x2 y

The v1 graph uses id1 and id2. Additional identifier-like columns available for richer specifications: id3.

Source Notes

Anonymized student-teacher data based on North Carolina Education Research Center records.

Valentin Verdier, 'Estimation and Inference for Linear Models with Two-Way Fixed Effects and Sparsely Matched Data'.

Historical Benchmark

2017 SEC benchmark timings for this dataset, in seconds:

MethodCitationSeconds
MAP-Aitken (Guimaraes 2012) 221.5
MAP-SD (Gaure 2013) 79.7
MAP-CG-Sym (Correia 2016) 61.7
MAP+Prune (Correia 2016) 116.6
LSMR (Gomez 2016) 132.0