Hello Statalist,
I'm trying to construct an innovation measurement from a paper: Corporate Innovations and Mergers and Acquisitions
The measurement is called Knowledge base overlap, it is documented in the paper appendix. Its calculation is: 1) find the set of patents that received
at least one citation from any of the acquirer’s patents with award years from ayr-3 to ayr-1 (“the acquirer’s knowledge base”)(ayr is the year of M&A), the set of patents that received at least one citation from any of the target firm’s patents awarded over the same three-year period (“the target firm’s
knowledge base”), and the intersection of these two sets as the set of patents cited by both the acquirer and the target firm (“the common knowledge base”). 2) compute the number of patents in “the common knowledge base".
I have a dataset with the M&A event year, the participating acquirer and target identifiers, and their patent number, I also have a dataset with the citing information of each patent, and I will merge them myself.
I know how to do it for each pair separately, and then loop through all pairs. But since I have lots of M&A firms, looping is like partitioning the records from one dataset into many datasets, with each of them containing only one pair. I wonder if there is a command to do this without looping?
Thank you so much
I'm trying to construct an innovation measurement from a paper: Corporate Innovations and Mergers and Acquisitions
The measurement is called Knowledge base overlap, it is documented in the paper appendix. Its calculation is: 1) find the set of patents that received
at least one citation from any of the acquirer’s patents with award years from ayr-3 to ayr-1 (“the acquirer’s knowledge base”)(ayr is the year of M&A), the set of patents that received at least one citation from any of the target firm’s patents awarded over the same three-year period (“the target firm’s
knowledge base”), and the intersection of these two sets as the set of patents cited by both the acquirer and the target firm (“the common knowledge base”). 2) compute the number of patents in “the common knowledge base".
I have a dataset with the M&A event year, the participating acquirer and target identifiers, and their patent number, I also have a dataset with the citing information of each patent, and I will merge them myself.
I know how to do it for each pair separately, and then loop through all pairs. But since I have lots of M&A firms, looping is like partitioning the records from one dataset into many datasets, with each of them containing only one pair. I wonder if there is a command to do this without looping?
Thank you so much
Comment