Thesis (Ph.D.)--University of Wollongong, 2004. / Typescript. Includes bibliographical references: p. 199-206.

Over the past decade of accounting and finance research, the Ohlson (1995) model has been widely adopted as a framework for stock price prediction. While using the accounting data of 391 companies from SP500 in this paper, Bayesian statistical techniques are adopted to enhance both the estimative and predictive qualities of the Ohlson model comparing to the classical approaches. Specifically, the classical methods are used for the exploratory data analysis and then the Bayesian strategies are applied using Markov chain Monte Carlo method in three stages: individual analysis for each company, grouping analysis for each group and adaptive analysis by pooling information across companies. The base data, which consist of 20 quarters' observations starting from the first quarter of 1998, are used to make inferences for the regression coefficients (or parameters), evaluate the model adequacy and predict the stock price for the first quarter of 2004, when the real observations are set as the test data to evaluate the predictive ability of the Ohlson model. The results are averaged within each specified group categorized via the general industrial classification (GIC). The empirical results show that classical models result in larger stock price prediction errors, more positively-biased predictions and have much smaller explanatory powers than Bayesian models. A few transformations of both classical and Bayesian models are also performed in this paper, however, transformations of the classical models do not outweigh the usefulness of applying Bayesian statistics.

Maps and models of seismic and tsunami risk are constructed from a variety of measurements taken in Indonesia, which have the potential to reduce loss of life and infrastructure. The first study uses the multichannel analysis of surface waves (MASW) method to calculate the time-averaged shear wave velocity to 30 m depth (Vs30). These measurements were taken at 58 sites in the city of Pacitan, Java and on the islands of Lombok, Ambon, and the Banda Islands. Vs30 calculations are compared with local geologic maps to extrapolate site class for locations not measured directly. Site class maps are then compared with Modified Mercalli Intensity (MMI) observations for three earthquake events that impacted Lombok and Ambon to identify regions where the MMI and Vs30 do and do not corroborate one another. Consistent with other Vs30 studies, the lowest values are observed on coastal alluvial plains and the highest values on steeper hillsides underlain by volcanic deposits. The second study focuses on a potential sector collapse of the volcano Banda Api within the Banda Islands. A field survey of its summit identified a steeply dipping normal fault striking NNE-SSW. This, along with the fissure geometry of the volcano's most recent eruption, reveals a failure plane along which a future sector collapse could occur. The numerical model Tsunami Squares (TS) predicts that the tsunami produced by this landslide would inundate an estimated 63% of buildings on the Banda Islands with waves as high as 82 m. These findings highlight the importance of installing a GPS receiver array on Banda Api to monitor the motion of its slopes. The third study analyzes sediment from trenches on the Banda Islands and Ambon to test if historical tsunamis that have impacted the area are preserved in the geological record. Potential tsunami deposits were identified by the presence of marine sand and larger clasts of marine carbonate in an environment which otherwise lacks large storms to bring such material onshore. Several dating methods constrain the ages of at least seven candidate tsunami deposits found in trenches onshore. One of these historical tsunamis (the event of November 26, 1852) is described in significant detail from several locations across the Banda Sea, which enables modeling of the event using a Bayesian statistical approach. The posterior of this model predicts the most likely epicenter was SW of Seram with a mean magnitude of Mw 8.8. It also makes other predictions about fault parameters. The region exhibits a marked slip deficit based on instrumental records of earthquakes in the area.