Accounting and Finance Project structure

Your Project should consist of the following chapters:
1. Introduction
2. Theoretical and Contextual Literature Review
3. Methodology
4. Data analysis and results
5. Conclusion
Developing a Project topic
• an aspect of accounting or finance that interests you
• something you have read, seen on TV or heard on radio
• link idea to potential future career / further study
Example of project ideas
• Governance role of board of directors
• Corporate social responsibility
• Governance and accountability with nongovernmental organisations (NGOs)
• Corporate Governance (UK, International, or
comparative, etc., whether it is about CG on its own
or linking it with firm performance, reporting, etc.)
• IFRSs of students’ choice, such as Segment Reporting
(IFRS8 Operating Segments), Intangible Asset
Reporting (IAS38 Intangible Assets)
• Voluntary disclosure in the areas of e.g. CSR,
environmental reporting, sustainability reporting
• Earnings management
• Financial statement analysis and firm valuation
• Directors pay and performance
• Directors pay and risk
• Corporate governance and performance
• Gender and pay
• Non-executives’ pay
• Share price anomalies: why are share prices higher
on certain days/weeks/months of the year?
• An empirical investigation of the impact of
economic fundamentals on the real exchange rate
• Determinants of bank profitability
• Determinants of capital structure
• Determinants of annual reports readability
• Tax avoidance
• Tax aggressiveness
Need to ensure:
• Ensure it is manageable focus your idea
• Ensure that:
• It is possible to measure / collect data for the dependent variable, otherwise you don’t
have a feasible topic
• You can measure / collect data most of the independent variables indicated by the
literature
• That you have a reasonable sample size (i.e. at least 30 observations)
Research Process
• Research Question• The ‘big question’, but don’t make it too big ensure it is a
single direct
question
• Sub-questions and / or objectives• Relate these to the detailed aspects of what
you will test the independent
variables of your model
• Hypotheses (to answer the sub-questions)• A series of statements that can be
made into questions with true/false type
answers to establish the ‘truth’
• The ‘research question’ is the ‘big question’ or ‘focus question’ of
your research…often used to phrase the dissertation title as a
question to cover the following ideas:
• WHAT ‘what are the determinants of…’
• WHERE in which country, industry…
• WHEN the time period of the study
• This has the advantage that it is easy to assess whether you have
done what you set out to do
Hypotheses are need to answer the sub-questions:
• These sum-up what you think in terms of the ‘research question’ and basis of
the question / sub-questions
• These should be statements that can be tested easily in terms of true/false
questions
• Therefore hypotheses should be worded so that you can test them:
• You should ask yourself is there a way to test it in terms of a statistical test?
• You should ask yourself can I get the data to test it?
• You will need to include data in different forms at different points in
your Project:
• To motivate your research Introduction
• To provide a context for your dissertation Introduction
• To test hypotheses Data analysis and results
• Think: when and why you might need to use secondary data
• Think: what must I know and examine in order to analyse my
question?
Data analysis: presentation, techniques and interpretation
Results presentation
• Don’t copy & paste SPSS output when you present your regression
results…create correct output tables in Word
• Consider how and when to use graphs in your empirical analysis
writing:
• Use line graphs to show trends
• Use X-Y graphs to show relationship between the two variables
• Disaggregate the data into different samples…and compare / contrast
across these different samples
Calculate across the whole sample…
• …or for disaggregated aspects of the data set for comparison:
• Mean
• Minimum
• Maximum
• Standard deviation
• Report any unusual patterns…and/or check for incorrectly inputted
data
Measures of association
• Explore a range of increasingly sophisticated measures of association
in your empirical analysis:
• Correlation shows the association between a pair of variables
• Simple regression again shows the association between a pair of
variables…and the impact that one variable has upon the other and elasticity
of the relationship
• Multiple regression shows the association between the dependent variable
(i.e. the exchange rate) and a range of independent variables…and the impact
that one variable has upon the other and elasticity of the relationship
Discussion of regression results
• Review the three ‘s’:
• Statistical significance
• Sign
• Size (and elasticity)
• Statistical significance:
• Review the statistical significance of the estimated coefficients t-test
• Review the statistical significance of the estimated model F-test
• Sign:
• Review the sign of the estimated coefficients in relation to:
• that expected/predicted by either economic theory…
• …or that found in previous empirical studies
• If your estimated coefficient’s sign is different to that expected/predicted…
• …you particularly need to discuss why this might be so
• Size:
• Review the predictive size of the model R2 and Adj. R2
• The goodness of fit (R2) shows how well your model performs…
• …but if you change the number of independent variables such that you want
to compare models…
• …you need to use the Adj. R2 which deflates the R2 according to the number
of independent variables
• Size:
• Review the size of the estimated coefficient…
• …note that you need to take into account the units of measurement of the
dependent and independent variable
• Review the elasticity of each independent variable at the sample mean