Research Methodologies Report: Finance, Company Bankruptcy Analysis

This report delves into research methodologies employed in financial analysis, specifically focusing on company bankruptcy prediction. The study utilizes data from the Bloomberg Terminal, extracting financial statements from both bankrupt and healthy US manufacturing organizations between 2008 and 2018. The research incorporates a sample of 50 companies and employs various techniques, including bootstrap tests and multi discriminant analysis. The methodology encompasses data collection, sample selection, and the application of group statistics to identify financial risks. The report outlines the use of bootstrap tests to assess predictor variable significance, especially with small sample sizes. Multi discriminant analysis is utilized to distinguish between bankrupt and non-bankrupt organizations, developing a model with coefficients for each variable to classify companies based on their scores. References include Kumar (2019), Mackey and Gass (2015), Quinlan et al. (2019), and Walliman (2017).

Running head: RESEARCH METHODOLOGIES
Research Methodologies
Name of the Student:
Name of the University:
Author’s Note:
Course ID:

1RESEARCH METHODOLOGIES
Table of Contents
1. Data section:................................................................................................................................2
2. Sample of failed and non-failed companies:...............................................................................2
3. Bootstrap test:..............................................................................................................................2
4. Group statistic:.............................................................................................................................3
5. Multi discriminant analysis:........................................................................................................3
References and Bibliopgraphy:........................................................................................................4

2RESEARCH METHODOLOGIES
1. Data section:
For this research, data have been gathered from Bloomberg Terminal owing to the fact
that the website provides real time analytics and equity trading, market and financial data for
active and non-active financial organisations. The financial statements of a list of the bankrupt
organisations in US and healthy organisations have been extracted and the financial statements
of the bankrupt organisations for the past two years before their bankruptcy have been gathered.
Moreover, for pertinent information, the annual reports of the organisations and the Edgar
Database have been used as well and the data accumulated have been analysed with the help of
Excel and SPSS.
2. Sample of failed and non-failed companies:
In the current research, focus has been kept on the period after the occurrence of the
financial crisis along with measuring cases of the previous ten years from 2008-2018. The actual
sample has been gathered from Bloomberg Bankruptcy Dashboard including 50 manufacturing
organisations, out of which 2 have become bankrupt and 24 are still alive in USA. The reason
behind the selection of this dashboard is that Bloomberg gives data on organisations with size of
above 500 and missing data have been found on smaller-sized organisations. The sample features
describe the particular areas, in which there has been application of discriminant function,
3. Bootstrap test:
The sample size chosen for this particular research is found to be small. This has
necessitated the need of viewing the significance effect on the predictor variables using small
sample of only 50 manufacturing organisations along with ascertaining whether the predictive

3RESEARCH METHODOLOGIES
power would vary with size. In case of Bootstrap test, the analysis is based on the confidence
interval of 95% and confidence interval type of “Bca”. In this context, Quinlan et al. (2019)
remarked that bootstrapping could be defined as any metric or test relying on random sampling
with replacement, as it allows allocating accuracy measures to sample estimates.
4. Group statistic:
Group statistic has assisted in defining each group and predictor variables for the
research. From the mean scores and standard deviation scores, bankrupt companies and alive
organisations, it becomes possible to identify the financial risk of the chosen organisations
(Kumar 2019). In addition, the group statistic has assisted in analysing the effectiveness of
certain ratios in a group analysis. Highly skewed implies to be unfavourable or favourable in
identifying discrimantive procedures. Moreover, the group statistic would help in identifying the
trend of each variable based on which the researcher has arrived at actual inference of the
research.
5. Multi discriminant analysis:
The multi discriminant analysis has helped in identifying the major differences between
bankrupt and non-bankrupt organisations by utilising the average of the variable ratios. By using
this model in the current research, the researcher has developed a group of coefficients, which
have been used in the form of weights of each variable (Mackey and Gass 2015). This would
help the US group organisations to be identified as bankrupt or not bankrupt depending on their
scores and identified cut off point.

4RESEARCH METHODOLOGIES
References and Bibliopgraphy:
Kumar, R., 2019. Research methodology: A step-by-step guide for beginners. Sage Publications
Limited.
Mackey, A. and Gass, S.M., 2015. Second language research: Methodology and design.
Routledge.
Quinlan, C., Babin, B., Carr, J. and Griffin, M., 2019. Business research methods. South Western
Cengage.
Walliman, N., 2017. Research methods: The basics. Routledge.