This paper studies the performance of alternative prediction models for conflict. The analysis contrasts the performance of conventional approaches based on predicted probabilities generated by binary response regressions and random forests with two unconventional classification algorithms. The unconventional algorithms are calibrated specifically to minimize a prediction loss function penalizing Type 1 and Type 2 errors: (1) an algorithm that selects...
See More
DETAILS
-
2017/05/26
-
Policy Research Working Paper
-
WPS8075
-
1
-
1
-
2017/05/26
-
Disclosed
-
Predicting conflict
-
Fragility
DOWNLOADS
COMPLETE REPORT
Official version of document (may contain signatures, etc)
*The text version is uncorrected OCR text and is included solely to benefit users with slow connectivity.
Citation
Celiku,Bledi; Kraay,Aart C..
Predicting conflict (English). Policy Research working paper|no. WPS 8075 Washington, D.C. : World Bank Group. http://documents.worldbank.org/curated/en/619721495821941125