Similar Stocks

다국어 초록 (Multilingual Abstract)

Similarity between two stocks is measured by the distance between their characteristics such as price, size, book-to-market, return on assets, and investment- to-assets. We find that after a stocks most similar stocks have experienced high (low) returns in the past month, this focal stock tends to earn an abnormally high (low) return in the current month. The long-short portfolio strategy sorted on similar-stocks past average return earns a monthly CAPM alpha of 1:25% and a Fama-French six- factor alpha of 0:85%. This similarity effect is robust after controlling for style investing and a wide range of well-known rm-level characteristics that can predict returns in the cross section. Our result is consistent with the increased propensity for investors to buy other stocks with similar characteristics after experiencing positive returns for a currently held stock. We also explore other potential explanations for our findings.

목차 (Table of Contents)

• Abstract
• 1 Introduction
• 2 Data Description and Summary Statistics
• 2.1 SIM Definition
• 2.2 Control Variables
• 2.3 Retail Order Imbalance
• 2.4 International Sample
• 3 Empirical Analysis: The Predictive Power of Similar-Stock Returns
• 3.1 Unconditional Sorts Based on the Average Returns of Similar Stocks
• 3.2 Double Sorting
• 3.3 Fama-MacBeth Regressions
• 3.4 Similarity Effect over the Long Horizon
• 3.5 Subsequent Earnings Announcements
• 3.6 Retail Order Imbalance
• 4 Further Robustness Checks and Alternative Channels
• 4.1 Subsample Analysis
• 4.2 Interaction with Institutional Ownership
• 4.3 Evidence from International Markets
• 5 Conclusions
• References
• Figure
• Table