JoseM Pavia -

JoseM Pavia

Full Professor

Jose M. Pavía, MSc in Maths and PhD in Economics and Business Science, is Quantitative Methods Professor in the Economics Faculty of the Universitat de Valencia, Spain. Pavia develops and applies statistical and machine learning methods in many areas of social sciences. With broad and varied research interests, his work focuses on the search for innovations that bridge the gap between theory and practical applications and include issues related to public policy evaluation, electoral processes, prediction, statistical (machine) learning, ecological inference, crime detection, survey research, sampling, public opinion, behavioral economics, experiments, regional economy, gender studies, or inequality. He is director of the Elections and Public Opinion Research Group of the Universitat de Valencia and Chair of Catedra Deblanc, for the application of statistical, economic and machine learning to detect money laundering and financial crimes. Pavía has made more than ten research stays in some of the most prestigious universities, including Stanford, Carlos III de Madrid, Leicester, Southtampton or Durham, and has leaded in dozens of research and transfer projects as IP. More info at .

Universitat de Valencia

Public Documents 1

Improving estimates accuracy of voter transitions

JoseM Pavia

and 1 more

June 07, 2022

The estimation of RxC ecological inference contingency tables from aggregate data defines one of the most salient and challenging problems in the field of quantitative social sciences. From the mathematical programming framework, this paper suggests a new direction for tackling this problem. For the first time in the literature, a procedure based on linear programming is proposed to attain estimates of local contingency tables. Based on this and the homogeneity hypothesis, we suggest two new ecological inference algorithms. These two new algorithms represent an important step forward in the ecological inference mathematical programming literature. In addition to generating estimates for local ecological inference contingency tables and amending the tendency to produce extreme transfer probability estimates previously observed in other mathematical programming procedures, they prove to be quite competitive and more accurate than the current linear programming baseline algorithm. The new algorithms place the linear programming approach once again in a prominent position in the ecological inference toolkit. We use a unique dataset with almost 500 elections, where the real transfer matrices are known, to assess their accuracy. Interested readers can easily use these new algorithms with the aid of the R package lphom.