Individual Equity Return Data From Thomson Datastream: Handle With Care!

Abstract

We compare individual U.S. equity return data from Thomson Datastream (TDS) with similar data from the Center for Research in Security Prices (CRSP) to evaluate TDS for use in studies involving large numbers of individual equities in markets outside the United States. We document important issues of coverage, classification, and data integrity and find that naive use of TDS data can have a large impact on economic inferences. We show that after careful screening of the TDS data, inferences drawn from TDS data are similar to those drawn from CRSP. We illustrate the importance of the screens we develop using U.S. TDS data by applying the screens to TDS data from four European equity markets.

Searching for a listed infrastructure asset class using mean–variance spanning

“…8 Setting a minimum infrastructure sector revenue threshold to 75 and 90% yields 650 and 554 stocks, respectively. 9 USD price and total returns are from Datastream using the methodology described in Ince and Porter (2006). 10 The firms thus identified comprise at most 12, 7, and 6.5% of the MSCI World market value as of December 31, 2014, for the 50, 75, and 90% revenue thresholds, respectively.…”

“…9 The number of firms identified as well as their geographic and industry distributions are available upon request. 10 Extreme monthly returns are identified following Ince and Porter (2006) and set to a missing value. Ince and Porter (2006) set an arbitrary cutoff of 300% for extreme monthly returns.…”

“…10 Extreme monthly returns are identified following Ince and Porter (2006) and set to a missing value. Ince and Porter (2006) set an arbitrary cutoff of 300% for extreme monthly returns. If R1 or Rt − 1 is greater than 300% and (1 + R1)/(1 + Rt − 1) − 1 is less than 50%, then R1 or Rt − 1 are set to missing.…”

Lead-Lag Relationships in International Stock Markets Revisited: Are They Exploitable?

“…When a firm stops trading, TRD repeats the last reported value for each datatype. Following Ince and Porter (2006), we handle this issue by setting to missing observations with a zero return, starting from the end of the sample. We also set all returns to missing that are affected by an extreme return reversal over two consecutive periods and returns larger than 800%.…”

Coasimetría idiosincrática y riesgo de insolvencia en el mercado de valores español

“…En segundo lugar, el resto de variables utilizadas, han sido extraídas de dicha base de datos con frecuencia mensual, considerando la variable tamaño como el logaritmo natural de la capitalización bursátil de cada empresa y siendo la ratio book-to-market el cociente entre el valor en libros y la capitalización de cada empresa. Después de aplicar varios filtros sugeridos en Ince and Porter (2006) para evitar errores derivados del uso de la base Thomson Financial, la muestra final ha quedado constituida por 212 empresas, con un mínimo de 70 al comienzo del periodo objeto de estudio y un máximo de 130 en 2007. Con objeto de disponer de una medida homogénea para todo el periodo considerado, como tipo de interés libre de riesgo se ha tomado la referencia de tipos de interés de convergencia de Maastrich (MCBY) publicados por EUROSTAT.…”