Survival analysis of companies’ delisting time in Indonesian stock exchange using Bayesian multiple-period logit approach
DOI:
https://doi.org/10.11113/mjfas.v13n4-1.864Keywords:
Survival, multiple-period logit, Bayesian, delisting, C-indexAbstract
Multiple-period logit model is equivalent to hazard model. This model is able to accommodate time varying predictor. In this work, the parameters of multiple-period model are estimated by using Bayesian inferences. There are three prior distributions used, i.e. improper uniform distribution, multivariate normal distribution, and Cauchy distribution. Criterion which is used to evaluate the proposed technique is C-index. The proposed method is applied to model the delisting time of companies listed in Indonesian Stock Exchange. The survival (delisting) time is driven by firm-specific predictors, i.e. financial ratios, that are calculated from quarterly financial report of companies in manufacturing sector span from the first quarter of 1990 until the third quarter of 2015. Two macroeconomic indicators are also considered as predictors. The empirical results show that the most appropriate prior is multivariate normal distribution. In addition, the proposed model is applied on windowing scheme by reducing the interval time as window in which the model estimator perform by its best.
References
Bender, R., Augustin, T., and Blettner, M. Generating Survival Times to Simulate Cox Proportional Hazards Models, Statistics in Medicine, 24, pp. 1713-1723, 2005.
Box, G. E. P. and Tiao, G. C., Bayeisan Inference in Statistical Analysis, New York: John Wiley and Sons, Inc., 1992.
Cole, R. A. and Wu, Q., Predicting Bank Failure using a Simple Dynamic Hazard Model, CFR Seminar Series Library, 2009.
Kleinbaum, D.G. and Klein , M. Survival Analysis : A Self-Learning Text, 3rd ed, London:Springer, 2012.
Gelman, A., Jakulian, A., Pittau, M. G., and Su, Y. S., A Weakly Informative Default Prior Distribution for Logical and other Regression Models, The Annals of Applied Statsitics, 2(4), pp. 1360-1383, 2008.
Mahjub, H., Faradmal, J., Goli, S. and Soltanian, A. R. Performance
Evaluation of Support Vector Regression Models for Survival Analysis: A Simulation Research, Internationa Journal of Advanced Computer Science and Applications (IJACSA), 7(6), pp. 381-389, 2016.
Haerdle, W. and Prastyo, D. D. Embedded Predictor Selection for Default Risk Calculation: A Southeast Asian Industry Study, in Chuen, D. L. K. and Gregoriou, G.N. (Eds.). Handbook of Asian Finance, Vol. 1, Financial Market and Sovereign Wealth Fund, pp. 131-148 San Diego: Academic Press, 2014.
Iriawan, N. Report on the Establishment of Early Warning System (EWS) of Corporate Bonds (Original Title : Laporan Hasil Pembentukan Early Warning System (EWS) Obligasi Korporasi), Research Report, Surabaya: ITS, 2005.
Ismail, N. M., Khalid, Z. M., and Ahmad, N. Estimating Proportional Hazards Model Using Frequentist and Bayesian Approaches, Malaysian Journal of Fundamental and Applied Sciences, 8(2), pp. 73-82, 2012.
Khoiri, H. A., Prastyo, D. D., and Purnami, S. W. A Simulation Study and Application of Feature Selection on Survival Least Square Support Vector Machines, The 13th International Conference on Mathematics, Statistics, and Their Applications (ICMSA), Kedah, 2017.
Prastyo, D. D. and Haerdle, W. K. Localising Forward Intensities for Multiperiod Corporate Default, SFB 649 Discussion Paper, 2014-040,2014.
Shumway, T. Forecasting bankruptcy more accurately: A simple hazard model, The Journal of Business, 74, pp. 101-124, 2001.
Suhartono, Rahmawati, D., Atok, R. M., Prastyo, D. D. and Ahmad, I. S., Determinant Factors of Publication in Scopus at ITS, The 12th International Conference on Mathematics, Statistics, and Their Applications (ICMSA), Banda Aceh, 2016, pp. 144-148.
Van Belle, V., Pelckmans, K., Suykens, J. A. and Van Huffel, S. Additive Survival Least-Squares Support Vector Machines, Statistics in Medicine, 29(2), pp. 296-308, 2010.
Van Belle, V., K. Pelckmans, J.A. Suykens, and S. Van Huffel, Support vector methods for survival analysis: a comparison between ranking and regression approaches, Artificial Intelligence in Medicine, 53(2), pp. 107- 118, 2011.
Wong, M. C. M., Lam, K. F. and Lo, E. C. M., Bayesian Analysis of Clustered Interval-Cencored Data, Journal of Dental Research, 84(9), pp. 817-821, 2005.
