Son Tran

Citation

Thesis Title
CLUSTER ANALYSIS AND FIRM PATTERNS OF EARNINGS PERSISTENCE:
A NEW APPROACH

Read article at Massey Research Online:

Mr Tran investigated a solution to the prevalent problem of heterogeneous-group specific coefficients (HGSC) within the context of the finance discipline. Specifically, he introduced a novel clustering procedure called regression oriented-weighted K-means clustering (ROWK). The performance of ROWK clustering was examined via both simulated and real data. Simulation results showed significant improvements from the adoption of ROWK relative to other standard clustering techniques. He further examined the performance of ROWK clustering using real data for earnings persistence models. ROWK outperformed other standard techniques in the sense of correctly identifying the underlying clusters on earnings persistence models. He also found that analysts’ forecasts only partially incorporate the information from cluster patterns in the short run, while ignoring impacts of these patterns on long-term future earnings. This research creates promising opportunities for future studies to apply ROWK clustering in other examined models when there is concern of HGSC.

Supervisors
Dr Jianguo Chen
Dr Carolyn Wirth