Contributions to financial econometrics and quantitative finance:PhD in three parts:Part 1: Liquidity transmission mechanism: Evidence from pre, during and post 2007 subprime crisis.Part 2: Benford's law tests on S&P 500 minute data.Part 3: Model and applications: IFBM-GARCH

This doctoral research consists of three parts; first part, discusses activities of international banks have been at the core of discussions on the causes and effects of the international financial crisis. Yet we know little about the actual magnitudes and mechanisms for transmission of liquidity shocks through international banks, including the reasons for heterogeneity in transmission across banks. The International Banking Research Network, established in 2012, brings together researchers from around the world with access to micro-level data on individual banks to analyse issues pertaining to global banks. This of research examines the linkages between market and funding liquidity pressures, as well as their interaction with solvency issues surrounding key financial institutions between pre/during and post crisis. This part of will undergo with the following steps: Step#1: test the significance of chosen data, analyse data feature, if series is non-stationary series, stabilize it by log difference or periodical difference. Step#2: ARMA model identification and parameter estimation: Identify the model and estimate parameter(s) according to series autocorrelation and partial correlation plot after stabilizing. Step#3: ARMA model test: Test model by statistical hypothesis testing method, if model is effective, then go to the fourth step, otherwise come back to the second step to adjust the model’s order and establish the model again. Step#4: ARCH effect test, model identification and parameter estimation: Do GARCH effect test for residual series, identify the model’s order, estimate the parameter, establish the ARMAGARCH model. Step#5: ARMA-GARCH model test: Test model by statistical hypothesis testing method, if model is effective, go to the sixth step, otherwise come back to the fourth step to adjust GARCH model’ order again. To select the order of GARCH model finally, we will also check the performance of that model on the validation set. Step#6: According to the established ARMA-GARCH model from steps 1-5 and integration to DCC-GARCH and BEKK-GARCH. Multivariate GARCH models have evolved from the optimal univariate GARCH model. The DCC specification will then allow the capture of possible structural breaks in the unconditional correlation amongst the variables. Finally, the BEKK-GARCH model will also be able to provide more detailed transmission information, apart from the conditional correlation. A multivariate GARCH model is estimated in order to test for the transmission of liquidity shocks across U.S. financial markets. It is found that the interaction between market and funding illiquidity increases/swerves sharply during the period of financial turbulence, and that bank solvency becomes important.
Second part describes an investigation into Benford’s Law for the distribution of leading digits in real data sets of S&P500 minutes volume and the corresponding log-returns over a long-time interval, [01/05/2013 - 29/12/2017], amounting to 481769 data points. This part of research addresses the frequencies of the first, second, and first two significant digits counts and explore the conformance to Benford's laws of these distributions at eleven different (117 days minute data points) levels of disaggregation of mathematical model of processes that might account for such a leading digit distribution have also been investigated. The log-returns are studied for either positive or negative cases. The results for the S&P500 minute volume data set are showing a huge lack of nonconformity. Such data sets have been examined and it was found that only a small fraction of them conform to the law. whatever the different levels of disaggregation. Some “first digits” and first two digits values are even missing. The causes of this non-conformity are discussed, pointing to the danger in taking Benford's laws for granted in huge data bases, whence drawing “definite conclusions”. Study found that based on the notion of taking the product of many random factors the most credible. This led to the identification of a class of lognormal distributions, those whose shape parameter exceeds 1, which satisfy Benford’s Law. This in turn led us to a novel explanation for the law: that it is fundamentally a consequence of the fact that many physical quantities cannot meaningfully take negative values. This enabled to develop a simple set of rules for determining whether a given data set is likely to conform to Benford’s Law. The agreements with Benford's laws are much better for the log-returns. Such a disparity in agreements finds an explanation in the data set itself: the inherent trend in the index. To further validate this, daily returns have been simulated calibrating the simulations with the observed data averages and tested against Benford's laws. One finds that not only the trend but also the standard deviation of the distributions are relevant parameters in concluding about conformity with Benford's laws.
Third part is to modify and test the IFBM GARCH methodology in terms of quantifying the impact of daily rates of return-on-investment activities in the observed S&P 500 stock index. The aim of the research, i.e., a special focus in the research, is to develop the modified GARCH methodology in the observed financial markets and to compare the obtained results between the variation of Modified GARCH models as well. The part of research is also aimed to study the performance of Modified IFBM GARCH models. A comprehensive empirical analysis of returns and conditional variances of the US stock exchange (S&P 500) index are conduct in order to estimate the GARCH models, and also the implementation of symmetric and asymmetric or (A-GARCH) models to observe the daily stock market volatility. Glosten (1993) GJR-GARCH model in term of an alternative edition of the asymmetric mode of Engel et al. (1990). For instance, it is often observed in GJR-GARCH model the asymmetric response is bounded by only the negative shocks of the market. Nelson’s (1991) E-GARCH for formulating the conditional variance equation to implement the method of ensuring, that the variance is positives., where the width of the research time horizon allows testing the modified GARCH methodology in the different periods. In addition to the use of modified IFBM GARCH econometric models, the focus of this work is to make use of existing, well-known Information Criteria (IC) to identify the stock index data-generating-process whenever the GARCH effect is present. Akaike Information’s Criteria (AIC) and Bayesian Information Criteria (BIC) have used for this experiment. Research provides different models with different parameter values and observed the abilities of information criterion in choosing the correct model from a given pool of models, as well as the appropriate tests that are suitable for and/or adapted to the specific characteristics of financial markets, examine irrational agent behaviour reacting to time dependent news on the log-returns for modifying a financial market evolution. The research results confirm the role and importance of the modified IFBM GARCH methodology for effective investment risk in financial markets, in this sense, the obtained research results will be useful to both the academic community and the professional public in the context of investment decision making.