Applying a "co-search" algorithm to Internet traffic at the SEC's EDGAR website, we develop a novel method for identifying economically-related peer firms and for measuring their relative importance. Our results show that firms appearing in chronologically adjacent searches by the same individual (Search-Based Peers or SBPs) are fundamentally similar on multiple dimensions. In direct tests, SBPs dominate GICS6 industry peers in explaining cross-sectional variations in base firms' out-of-sample: (a) stock returns, (b) valuation multiples, (c) growth rates, (d) R&D expenditures, (e) leverage, and (f) profitability ratios. We show that SBPs are not constrained by standard industry classification, and are more dynamic, pliable, and concentrated. We also show that co-search intensity captures the degree of similarity between firms. Our results highlight the potential of the collective wisdom of investors -extracted from co-search patterns -in addressing long-standing benchmarking problems in finance.JEL: D83, G0, M2
Introduction: About one-third of breast and ovarian carcinoma patients have circulating antibodies reactive with polymorphic epithelial mucin (MUC1), either free or bound to immune complexes. While the presence of these immune complexes has prognostic significance in breast cancer patients, the significance of free MUC1 antibodies is less clear. The objective of this study was to develop a reliable assay for the accurate determination of circulating free antibodies to MUC1. Material and Methods: We developed an enzyme-linked immunosorbent assay (ELISA) (PEM.CIg) employing a 60 mer peptide (a triple tandem repeat sequence of the MUC1 peptide core) conjugated to bovine serum albumin and peroxidase-labeled antihuman immunoglobulin G or M antibodies. The assay was standardized and its analytical performance evaluated. A total of 492 serum samples were obtained from 40 healthy men, from 201 healthy women (including 55 women without a history of pregnancy and 45 pregnant women), and (before primary treatment) from 62 benign breast tumor patients and 190 breast cancer patients. MUC1 serum levels were determined with commercial CA 15-3 tests. Results: Circulating antibodies to MUC1 are present both in healthy subjects and in breast cancer patients. The within- and between-assay coefficients of variation were, respectively, 2 and 12% for the IgG determinations and 1.2 and 3% for the IgM determinations. Correlation coefficients for serially diluted serum samples ranged from 0.9998 to 0.9920 for IgG and from 0.9996 to 0.9818 for IgM determinations. The reactivity of serum samples was partially blocked by the addition of various MUC1 peptides and by MUC1 mucin. The inhibiting effect of modified 60 mer peptides suggests the presence of antibodies directed to more than one epitope. Conclusions: The PEM. CIg assay is a reliable ELISA for measuring free MUC1 antibodies in serum. We are in the process of relating the results obtained in the breast cancer group to disease outcome to evaluate its prognostic significance. In addition, the assay may become a useful tool for vaccine therapy monitoring.
We compare the performance of a comprehensive set of alternative peer identification schemes used in economic benchmarking. Our results show the peer firms identified from aggregation of informed agents' revealed choices in Lee, Ma, and Wang (2014) perform best, followed by peers with the highest overlap in analyst coverage, in explaining cross-sectional variations in base firms' out-of-sample: (a) stock returns, (b) valuation multiples, (c) growth rates, (d) R&D expenditures, (e) leverage, and (f) profitability ratios. Conversely, peers firms identified by Google and Yahoo Finance, as well as product market competitors gleaned from 10-K disclosures, turned in consistently worse performances. We contextualize these results in a simple model that predicts when information aggregation across heterogeneously informed individuals is likely to lead to improvements in dealing with the problem of economic benchmarking. JEL: D83, G11Keywords: peer firm, benchmarking, EDGAR search traffic, co-search, analyst coverage, industry classification, wisdom of crowds * The authors can be contacted at clee8@stanford.edu, paulma@umn.edu, and charles.cy.wang@hbs.edu. We thank Boris Groysberg, Paul Healy, Ryan Buell, Kai Du, Akash Chattopadhyay, Andrew Jing Liu, Daniel Malter, Tatiana Sandino, Pian Shu, Martin Szydlowski, Akhmed Umyarov, and Aaron Yoon for helpful comments and suggestions. We are very grateful to Scott Bauguess at the Securities and Exchange Commission for assistance with the EDGAR search traffic data. We also thank Kyle Thomas for excellent research assistance. All errors remain our own.
Most research on population sex imbalance in China has focused on the One-Child Policy era. However, because much of China's fertility decline occurred during the 1970s, we investigate the possibility that sex ratios began rising during this period (as predicted by theory) before the One-Child Policy. RESULTS Analyzing sex ratios between 1960 and 1987 by birth order and sibship sex composition, we find that among the subset of couples expected to have the greatest demand for sons (those at higher parities without previous sons), sex ratios at birth reached 115-121 boys per 100 girls during the 1970s-implying approximately 840,000 to 1,100,000 girls missing from Chinese birth cohorts during the 1970s. Importantly, these results do not appear to be driven by differential under-reporting of living girls, or instances of adoption. Given the absence of ultrasound technologies prior to 1979, they imply the presence of postnatal sex selection in China during the 1970s. CONTRIBUTION Our work makes several important contributions to existing literature. First, we focus on the subset of couples among whom the demand for sons is predicted to be the strongest: higher parity couples not yet having a boy. Second, we estimate sex ratios by single year of age (from birth to age 4), distinguishing differential rates of infant death from more gradual neglect of girls as they age throughout childhood. Third, we
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.