4,249 works (including 483 books, 3,406 articles and 357 historical documents), 150 years of economic thought. Each one summarized and searchable.
What information disappears when many economic relations become a single macroeconomic model? In this essay, Gerhard Tintner and A. Sondermann make that loss an explicit statistical problem: a reduced model is judged by how well it reconstructs individual outcomes, under specified weights and probability distributions. Their distinctive emphasis is on optimal approximation when exact aggregation fails—not on assuming that totals preserve the relations beneath them. Generalized inverses and eigenvector methods supply mathematical tools, while consumption examples show why differences among individuals and their association with income can alter aggregate coefficients. Readers can discover both the possibilities and the limits of statistical reduction: preserving dominant variation, minimizing reconstruction error, and reproducing economic structure are different objectives, and need not yield the same aggregate.