Matrix Completion of World Trade

  title={Matrix Completion of World Trade},
  author={Giorgio Gnecco and Federico Nutarelli and Massimo Riccaboni},
This work applies Matrix Completion (MC) – a class of machine-learning methods commonly used in the context of recommendation systems – to analyze economic complexity. MC is applied to reconstruct the Revealed Comparative Advantage (RCA) matrix, whose elements express the relative advantage of countries in given classes of products, as evidenced by yearly trade flows. A high-accuracy binary classifier is derived from the MC application, with the aim of discriminating between elements of the… Expand


A New Metrics for Countries' Fitness and Products' Complexity
It is shown that a non-linear iteration is necessary to bound the complexity of products by the fitness of the less competitive countries exporting them, and the correct and simplest approach to measure the competitiveness of countries is the one presented in this work. Expand
The building blocks of economic complexity
It is shown that it is possible to quantify the complexity of a country's economy by characterizing the structure of this bipartite network in which countries are connected to the products they export, and that deviations from this relationship are predictive of future growth. Expand
Reconciling contrasting views on economic complexity
The authors recast both methods in a mathematical and multidimensional framework to reconcile both and show that the conflicts between the two methodologies to measure economic complexity can be resolved by a neat mathematical method based on linear-algebra tools within a bipartite-networks framework. Expand
De-biasing low-rank projection for matrix completion
It is shown that under non-uniform deterministic sampling, this method yields a biased solution, and a proposed algorithm to de-bias it is proposed, and an alternative which is based on projection onto the max-norm ball whose robustness to noise tolerates arbitrarily non- uniform sampling is suggested. Expand
Matrix completion and low-rank SVD via fast alternating least squares
This article develops a software package softlmpute in R for implementing the two approaches for large matrix factorization and completion, and develops a distributed version for very large matrices using the Spark cluster programming environment. Expand
Missing Not at Random in Matrix Completion: The Effectiveness of Estimating Missingness Probabilities Under a Low Nuclear Norm Assumption
This paper tackles MNAR matrix completion by solving a different matrix completion problem first that recovers missingness probabilities, and establishes finite-sample error bounds for how accurate these probability estimates are and how well these estimates debias standard matrix completion losses for the original matrix to be completed. Expand
A Singular Value Thresholding Algorithm for Matrix Completion
This paper develops a simple first-order and easy-to-implement algorithm that is extremely efficient at addressing problems in which the optimal solution has low rank, and develops a framework in which one can understand these algorithms in terms of well-known Lagrange multiplier algorithms. Expand
Spectral Regularization Algorithms for Learning Large Incomplete Matrices
Using the nuclear norm as a regularizer, the algorithm Soft-Impute iteratively replaces the missing elements with those obtained from a soft-thresholded SVD in a sequence of regularized low-rank solutions for large-scale matrix completion problems. Expand
Statistical Learning with Sparsity: The Lasso and Generalizations
Statistical Learning with Sparsity: The Lasso and Generalizations presents methods that exploit sparsity to help recover the underlying signal in a set of data and extract useful and reproducible patterns from big datasets. Expand
A change of perspective in network centrality
A new perspective is proposed where centrality definition arises naturally from the most basic feature of a network, its adjacency matrix, and different centrality measures naturally emerge, including degree, eigenvector, and hub-authority centrality. Expand