Cemil Kuzey & Victor Raj (Murray State)

Project:

I plan to incorporate diverse data analysis approaches such as univariate, multivariate, supervised, and unsupervised machine learning approaches in my research projects. Initially, data preprocessing with missing data analysis, winsorization, imputation of the missing values and the detection of outliers will be performed. After the data screening process, univariate analysis with descriptive statistics will be used. Factor Analysis with principal component analysis (PCA) will be employed to obtain independent test variables. In addition, correlation analysis, time-fixed effect regression analysis for panel data, and moderation analysis with visual illustrations are performed for the investigation of baseline models. Furthermore, the robustness of the baseline analysis results is investigated using moderation analysis with alternative independent test variables. The sample includes country-year, firm-year panel data for the research project with a sample size of approximately 10,000 observations. During the moderation analysis, a threshold test will be used to find the optimum cut-off value for creating the sub-samples.

Resource request on Openstack

Software

1) IBM SPSS Statistics V25

2) IBM SPSS Modeler V18.2

3) EView 11

4) RStudio +R

5) Anaconda3 with Python

6) Rapid Miner (Free)

8) Tableau-2020.3

8) MPlus over Remote Desktop