We argue that comprehensive out-of-sample (OOS) evaluation using statistical decision theory (SDT) should replace the current practice of K-fold and Common Task Framework validation in machine learning (ML) research. SDT provides a formal framework for performing comprehensive OOS evaluation...

We introduce the concept of a triad census of a digraph arid show how it can be used to enumerate various types of subgraph configurations. We give the basic probabilities needed for computing means and variances for a triad census under the U-MAN distribution for digraphs. These concepts are...

This review article, which was solicited by the Geneva Risk and Insurance Review, surveys work that has been done using an empirical framework for analyzing selection in insurance markets developed by Einav, Finkelstein, and Cullen (2010). We briefly review that framework, and then describe a...

Since 1979, the Bureau of Labor Statistics (BLS) has nearly quadrupled the size of the sample used to estimate monthly employment changes. Although first-reported employment estimates are still noisy, the magnitude of sampling variability has declined in proportion to the increase in the sample...

Analysts using the Health and Retirement Survey (HRS) often require information on earnings, labor market attachment, and social security benefits in order to better understand the factors affecting retirement and well-being at older ages. To this end, several derived variables were constructed...

This paper applies the nationality-adjusted net sales and value added framework proposed in Baldwin and Kimura (1996) to Japan. Despite possibly large estimation errors due to statistical deficiencies, the framework is very useful for analyzing the relationship of the Japanese economy to the...

The paper considers the problem of choosing a multilateral system of index numbers in order to make aggregate price and quantity comparisons between many countries and regions. The problems involved in choosing functional forms in order to make these comparisons are both theoretically and...

Despite the impression that the federal statistical system has been starved for resources, the record over the last decade or so, judged by conventional deflation methods, has been one of rough stability. The quality of U.S. international data on commodity and service trade, direct investment,...

This paper examines the properties and prevalence of measurement error in longitudinal earnings data. The analysis compares Current Population Survey data to administrative Social Security payroll tax records for a sample of heads of households over two years. In contrast. to the typically...

This paper combines information from two sources of U.S. private payroll employment to increase the accuracy of real-time measurement of the labor market. The sources are the Current Employment Statistics (CES) from BLS and microdata from the payroll processing firm ADP. We briefly describe the...