Text-Based Measures of Firm Position and Differentiation in Technology Space

Creators:

Sam Arts; Bruno Cassiman; Jianan Hou (built on the DISCERN database by Ashish Arora, Sharon Belenzon & Lia Sheer)

Publication Date:

2023

Data Category:

Dataset Description:

Data and code that represent each firm–year technology portfolio as a vector built from the stemmed technical keywords in the firm's patents (title, abstract, claims). Cosine similarity between vectors gives a text-based measure of how similar two firms are technologically, and of how differentiated a firm is. Built on DISCERN, which links roughly 1.35 million U.S. patents (1980–2015) to Compustat firms.

Variables:

Variable name	Variable definition
Stemmed technical keywords	Unique cleaned/stemmed keywords extracted from the title, abstract, and claims of a firm's patents (the text layer preserved in the deposit).
Tech similarity (text)	Cosine similarity between two firms' patent-text keyword vectors in a given year.
Technology differentiation	Extent to which a firm's technology portfolio differs from those of other firms, based on the text-based similarity measure.
Number of patents	Count of patents owned by the firm with a filing year in the relevant window.
Citation-weighted patents	Patent count weighted by forward citations.
Tech similarity (class / subclass)	Comparison measures computed the same way but using USPC patent classes / subclasses instead of patent text.
Tobin's Q / firm performance	Firm market value and profitability outcomes (Compustat) related to technology differentiation.

Details:

Bookmark this Dataset/Publication

Text-Based Measures of Firm Position and Differentiation in Technology Space

U.S.-Mexico Border Surveillance Data

Replication Dataset for Patent Publication and Innovation

Research Data and Replication Files (Expected Returns of Anomalies)

Text-Based Measures of Firm Position and Differentiation in Technology Space

Sign In

Register

Reset Password