Rosmanjawati Abdul Rahman

Selecting a Good Stochastic System for the Large Number of Alternatives

In this article, we present the problem of selecting a good stochastic system with high probability and minimum total simulation cost when the number of alternatives is very large. We propose a sequential approach that starts with the Ordinal Optimization procedure to select a subset that overlaps with the set of the actual best m% systems with high probability. Then we use Optimal Computing Budget Allocation to allocate the available computing budget in a way that maximizes the Probability of Correct Selection. This is followed by a Subset Selection procedure to get a smaller subset that contains the best system among the subset that is selected before. Finally, the Indifference-Zone procedure is used to select the best system among the survivors in the previous stage. The numerical test involved with all these procedures shows the results for selecting a good stochastic system with high probability and a minimum number of simulation samples, when the number of alternatives is large. The results also show that the proposed approach is able to identify a good system in a very short simulation time.

Finite mixture model: a comparison method for nonlinear time series data

Global Financial Crisis 2009 caused the finance collapse in Asia countries. The crisis leads to the economic and financial time series confronted with structural changes or jumps during the time. Stock price and exchange rate play a crucial role as symbol of development for a countrys economy. Hence, it is critical to investigate the relationship between these two variables. A two-component finite mixture model is used to examine this issue. In spite of that, maximum likelihood estimation and Bayesian method are applied to fit the finite mixture model. Additionally, this paper concerns on investigating the developing Asian countries since they are easily affected by economic changes of the western countries. In relation to the results, there is a positive relation between the exchange rate and stock price for Malaysia, Thailand, Philippines and Indonesia. Also, this paper highlighted that Bayesian method is superior to maximum likelihood in modelling nonlinear time series data although both methods provided roughly equal results.

Discriminant Analysis: A Study on Corporate Distress

The main purpose of this study is to identify which financial ratios are significantly important in predicting distressed companies. This study applied the discriminant analysis on a sample of 28 distressed and 28 healthy firms listed in the Bursa Malaysia during the period from the year 2004 to 2008. The distressed companies represented by Practice Note 17 companies (PN17) and the healthy companies are matched belonged to the same industries classification and have the closest assets. Through the usage of multivariate test, it is found that the financial ratios are able to discriminate the two groups (healthy and distress situation). The model appeared to be fairly accurate with classification accuracy rate more than 70 % up to 3 years before distressed. In addition, Return on Assets is the most important predictor of financial distress for three consecutive years.

Sequential selection approach for selecting the best system

Consider the problem of selecting the best simulated system with high probability, from a finite and huge set of alternative systems. The best system might be the one that has the maximum or minimum performance measure. In this paper, we present a sequential method that uses the Ordinal Optimization procedure to select randomly a subset that overlaps with the set of the actual best m% systems with high probability from the search space. The next step, we use Optimal Computing Budget Allocation technique to allocate the available computing budget in a way that maximizes the probability of correct selection. This follows by a Subset Selection procedure to get a smaller subset that contains the best system from the subset that is selected before. Finally, we use the Indifference-Zone procedure to select the best system among the survivors in the previous stage. The results of the empirical experiments show that this approach selects the best simulated system with high probability and a minimum number of simulation replication, when the number of alternatives is huge.