Page 257 - Contributed Paper Session (CPS) - Volume 3
P. 257

CPS2011 Dominique H. et al.
            conservative and bold financing strategies are adopted. It is interesting to note
            that the companies within a cluster present a heterogeneous mix of business
            content and industry. This means that the temporal dynamics of companies’
            financial  performance  seems  not  determined  by  traditional  classification
            indicators, such as by industry or line of business, and that these traditional
            classification indicators are insufficient to underpin an efficient comparison
            among companies.

                      Figure 4: Debt ratio constructs time series in each cluster





























                In terms of temporal dynamics, it seems that the majority of companies
            seek stable financial leverage in the long-run (see Figure 4); the fluctuation in
            the  short  run  may  reflect  the  inefficiency  in  the  process  of  debt  issuance.
            However, we identify two clusters (cluster 2 and cluster 13) that act differently.
            Cluster 2 shows a strong upward momentum at the beginning and end stages,
            while flat in the middle stage. Cluster 13 shows a downward trend during the
            first half and gradually becomes flat during the second half. Further research
            regarding the particular companies in these two clusters indicates that these
            unique behaviours of financial leverage signal important changes within the
            firms.  For  example,  the  excessive  leverage  showed  in  cluster  2  is  likely
            associated  with  Home  Depot’s  aggressive  engagement  in  buybacks  in  the
            stock market; the debt behaviour in cluster 13 may be highly associated with
            MHK’s acquisition in 2005 which is financed with debt.
                Under the framework of SOM, the comparison can be done by treating
            prototypes (Figure 5) as benchmarks.



                                                               246 | I S I   W S C   2 0 1 9
   252   253   254   255   256   257   258   259   260   261   262