Social and Behavioral Sciences Commons^™

Essays On Oil Price Volatility And Irreversible Investment, Daniel Joseph Pastor

Wayne State University Dissertations

In chapter 1, we provide an extensive and systematic evaluation of the relative

forecasting performance of several models for the volatility of daily spot

crude oil prices. Empirical research over the past decades has uncovered

significant gains in forecasting performance of Markov Switching GARCH

models over GARCH models for the volatility of financial assets and crude

oil futures. We find that, for spot oil price returns, non-switching models

perform better in the short run, whereas switching models tend to do better

at longer horizons.

In chapter 2, I investigate the impact of volatility on firms' irreversible investment decisions using real …

Distribution-Free Trends Test To Determine The Construct Validity Of An Anti-Social Criminal Attitudes Scale, Holly Ann Child

Wayne State University Dissertations

The Sawilosky's I-Test was developed to as an alternative method to evaluate construct validity, more specifically, in regards to the Multitrait-Multimethod Matrix designed by Campbell and Fiske (1959). Typically, researchers use a method by Campbell and Fiske that involves a subjective “physical” look at the matrix to determine validity. Sawilowsky’s I-Test offers a statistical approach that incorporates the current practice but removes the subjectivity involved in this process.

There are only two existing studies that look at the I-Test, Sawilowsky in 2002 and Cuzzocrea in 2007. Both studies found that although the I-Test is not a perfect statistic, it provides …

Novel Machine Learning Methods For Modeling Time-To-Event Data, Bhanukiran Vinzamuri

Wayne State University Dissertations

Predicting time-to-event from longitudinal data where different events occur at different time points is an extremely important problem in several domains such as healthcare, economics, social networks and seismology, to name a few. A unique challenge in this problem involves building predictive models from right censored data (also called as survival data). This is a phenomenon where instances whose event of interest are not yet observed within a given observation time window and are considered to be right censored. Effective models for predicting time-to-event labels from such right censored data with good accuracy can have a significant impact in these …