No announcement yet.
  • Filter
  • Time
  • Show
Clear All
new posts

  • making a comparable portfolio of peers for each observation


    I am currently writing my Msc. Thesis in Finance on the topic of ''the Private Company Discount''. In order to research this phenomenon I have collected a dataset of approx. 2000 private M&A transactions and 6000 public M&A transactions from 1985-2017. In order to calculate ''the discount'' I want to make a reference portfolio of public transactions for each private transaction observation. I can then compare the median deal multiple of the public transactions with the private transaction to establish the Private Company Discount.

    However, I am having trouble on how to construct the reference portfolio of public transactions for comparison. I have three criteria on which the public transaction should match with the private transaction:
    1. the announcement date of the public transaction must lie within a range of three calendar years before or after the announcement date of the private transaction
    2. at least the first 2 digits of the industry SIC code must be the same, but matching on the 3 or 4 digit level is preferred over the 2-digit level
    3. The companies must be similar in size such that Total assets or net sales must be similar

    for criteria 3 I have already divided the dataset in percentiles such that I have 4 categories of ''size'' for these 2 variables. So the transaction is matched if either Total assets ór net sales lies in the percentile (or both).

    I want to establish a reference portfolio of at least 5 comparable transactions per observation, whenever this cannot be established the year criterion can be relaxed.

    I would not know how I should establish this with Stata, any help would be very much appreciated! I am using Stata 14 on Windows10

    Thanks in advance!

  • #2
    We can understand your dataset only to the extent that you explain it clearly. The best way to explain it is to show an example. The community-contributed command dataex makes it easy to give simple example datasets in postings. See help dataex for details (see faq). In addition, it would be informative if you could also show an example of your desired dataset.


    • #3
      Thank you for responding. My problem is actually almost identical to a previous post on this forum:

      Only unfortunately there is no solution posted to this problem. Would anybody be able to help with this?


      • #4
        does anybody know a method to help me with this? I really need a solution!


        • #5
          You might want to take a look at
          ssc install rangejoin
          help rangejoin