Study on Mergers & Acquisitions Target Selection Based on Self-Organizing Mapping Neural Network

Applying the Negative Selection Algorithm for Merger and Acquisition Target Identification Theory and Case Study

“…Cheh, Weinberg and Yook [18] use a sample space of 1275 unacquired companies and 173 acquired companies. Hongjiu, Yanrong, and Shufen [20] use a set of ten enterprises to distinguish between target and nontarget companies. Little, Hickey, and Brabazon [21] use a total of 200 quoted US to separate and cluster the merged and nonmerger company.…”

“…Since the acquisition process starts after July 2008, the period of analysis is considered as four years from 2005 to 2008. Based upon prior literature [10] [16] [17] [20] [21] [38], a broad list of 18 financial variables are chosen as features to describe the self and nonself banks (check features in Table1).…”

“…The type II error of the DA is 0.4355, compared to the type II error of ANN (0.5860 -0.8763), which indicates that compared to ANN, DA is better at classifying companies as not acquired when in practice they are acquired [19]. Hongjiu, Yanrong, and Shufen [20] and Little, Hickey, and Brabazon [21] distinguish between target and non-target companies by using SOM. Although the study in [20] does not provide an accurate value of their classification accuracy, the authors claim that their results are in accordance with the expert judgment of M&A target selection.…”

“…Hongjiu, Yanrong, and Shufen [20] and Little, Hickey, and Brabazon [21] distinguish between target and non-target companies by using SOM. Although the study in [20] does not provide an accurate value of their classification accuracy, the authors claim that their results are in accordance with the expert judgment of M&A target selection. In [21], the authors achieve an in-sample accuracy of 94.80% for 100 companies that are successfully acquired; and an out-of-sample accuracy of 95.20% for 100 companies that are not acquired.…”

“…For the acquiring company, the choice of a correct set of indicator variables (features) may go a long way in accurately identifying its compatible target. Ideally the feature set should consist of indicator variables from the company's past and present production, financial, marketing, and human resource conditions; and also the industry and other macroeconomic conditions [20]. However, there are no hard and fast rules for such choices as they depend on the judgment of the acquiring company's M&A experts and the availability of data of both the acquirer and the potential takeover targets.…”