STAT2000 Quantitative Analysis
ASSESSMENT 1 - HOMEWORK QUESTIONS Module 1, 2, 3, 4 Homework Questions Upload your responses in a MS Excel or MS Word file for the homework questions in this module. Course Name: STAT2000 Quantitative Analysis Book : Lee, C., Lee, J. C., & Lee, A. C. (2013). Statistics for Business and Financial Economics (3rd ed.). New York, NY: Springer.
- Weekly practice questions in each module will be graded for a total of 20% .
- Chapter 1: Questions and problems 1, 5, 6, 7, 8, 11, 14, 16, 22, 30
- Chapter 2: Questions and problems 9, 14, 17, 19, 21, 37
- Chapter 3: Questions and problems 9, 10, 12, 20 Additional: Download data from the World Bank website ( http:// data.worldbank.org/data -‐ catalog/world -‐ development -‐ indicators ) to replicate Table 2.2 of the text book with Australian data.
- Chapter 4 : Questions and problems 1, 6, 7, 10, 13, 21, 28(a), 33, 42, 63
- Chapter 5 : Questions and problems 18, 23, 37, 43, 65, 72, 76, 79, 84, 85, 86
- Question 6: Questions and problems 1, 2, 3, 4, 5, 6, 7, 15, 28, 29, 57
- Question 8: Questions and problems 5, 6, 7, 8, 10, 11, 12, 14, 15, 22
- Question 11: Questions and problems 1, 3, 4, 6, 8, 10, 11, 12, 13, 15, 16
- Chapter 14: Questions and problems 23 -‐ 30, 33, 34, 40
- Answer the homework questions and save your answers in an MS Excel file
- Upload your submission using the upload feature in Blackboard for review by your learning facilitator
Assessment Group Assignment
Context: Research topic: Economic growth, private credit and remittances Objective: To examine the relation of economic growth with private credit and remittances Context: International Data period: 1996 -‐ 2012 Data source: World Development Indicators and World Governance Indicators
1. Describe the data (e.g., mean, standard deviation, skewness, kurtosis, 1st quartile, median, 3rd quartile, max, and min)
2. Prepare correlation matrix
3. Report regression results
4. Write a report stating clearly the objectives of this research, background (i.e., the importance of this research), the hypothesis, analysis of your findings, and the conclusion
We apply multiple regression analysis to estimate the following baseline model.
Growth = β0 + β1 × Privatecredit + β2 × Remittances + β3 × Manufacturing + β4 × Services + β5 × Foreigndirectinvestment + β6 × Workforceparticipation + β7 × Wages + β8 × Inflation + β9 × Inequality + β10 × Income+ β11 × Income2 + β12 × Controlofcorruption + β13 × Stability + ε
where Growth is the dependent variable. Private credit and Remittances are the research variables. Manufacturing, Services, Foreign direct investment, Workforce participation, Wages, Inflation, Inequality, Income, Income 2, Control of corruption and Stability are the control variables. The variables are discussed in Table 1.
Table 1: Variable Descriptions
|Growth||Annual growth of real GDP per capita||WDI|
|Privatecredit||Domestic credit to private sector (% GDP)||WDI|
|Remittances||Personal remittances, received (% GDP)||WDI|
|Manufacturing||Manufacturing, value added (% GDP)||WDI|
|Services||Services, value added (% GDP)||WDI|
|Foreigndirectinvestmen t||Foreign direct investment, net inflows (% GDP)||WDI|
|Workforceparticipation||Labor force/Population aged over 15 years||WDI|
|Wages||Wages and salaried workers, total (% of total employed||WDI|
|Inflation||Consumer price inflation||WDI|
|Income||Real GDP per capita in the host country (1000s of constant 2005 international dollar)||WDI|
|Income2||Squared term of Income||WDI|
|Controlofcorruption||Index: Control of corruption (-‐2.5 to 2.5)||WGI|
|Stability||Index: Political stability and Absence of violence/terrorism (-‐2.5 to 2.5)||WGI|
Structure of a quantitative research paper
We generally find the following structure in quantitative research papers.
Non -‐ numbered: Abstract
- Literature review and hypotheses development
- Findings and analysis (or discussion)
An abstract is a short summary (e.g., 150 words) of a research article.
This section discusses the background and motivation of the stud y. Whilst the primary motivation for academic research comes from the gap in the literature, the need for the study should be made clear. Unless there are some perceived benefits and practical (or policy) implications, there is no need to do a research. On ce the importance of the research topic is explained and the gap in the literature is identified, the research question (s) are raised in this section. There might be variation to writing the research questions. Some research mentions the research aims (or objectives) instead of formulating the research questions. Both approaches are very common and accepted.
2. Literature review and hypotheses development
The literature review and hypotheses development section formulates the testable hypotheses to answer the research questions introduced in Section 1. Hypotheses are developed ex -‐ ante (before the data collection) and so it is generally advised that these should be based on the theory or intuitive reasoning rather than grounded on other empirical evidence. It is also suggested to have directional (e.g., positively or negatively related) hypotheses rather than non -‐ directional (e.g., related). Non -‐ directional hypotheses are considered exploratory, for example where theory cannot provide an unambiguous expectat ion, and so it is better to put effort into developing an ex -‐ ante prediction.
We need to explain the way the hypotheses will be tested. This section covers the data, model and estimation procedure.
Sampling and data sources are covered under data.
3.2. Model and estimation procedures
A multiple regression model may appear as follows:
Dependent variable = Intercept + Coefficient (s) × Independent variable (s) + Coefficient (s) × Control variable (s) + Error term Yi = β0 + βi Xi + βi Zi + ε
The full form model expression (spelling out each variable) depends on the data structure. For example, if two variables on the right hand side of the equation are highly correlated, we may include only one of these two variables. For this reason, descriptive statistics are provided in the data section.
The estimation procedure depends on a number of factors including the types of variables. For example, if the dependent variable is continuous (i.e., can take on any value within a limit), an ordinary least square regression (OLS) estimation might be suitable. If the dependent variable is dichotomous (i.e., takes on a value of either 1 or 0), a logistic regression (logit) estimation might be appropriate.
This section presents the results of the estimation. Generally, the regression outputs are reported in this section. The Regression analysis comments on the model fit, explanatory power and statistically significant variables. It also presents additional analyses for robustness.
5. Findings and analysis
Findings of the study are discussed in this section. The discussion involves interpreting the statistical output. More specifically, the discussions should be for a general audience as well as experts in the field. So a non -‐ technical ex planation is always useful for a wider audience. The findings are also evaluated by comparing with similar previous studies to position this study in the literature. Economic interpretations (if this can be done) also improve the quality of the research. F or example, an increase of $1,000 in payment for attending meeting reduces the meeting non -‐ attendance problem of directors.
This section sums up the study. For a reader who is too busy to read the whole paper, this section should provide a summary of the whole paper. It should touch base on the research questions, the hypotheses, the test results, major findings, practical and policy implications, and the contribution of the study.
Disclaimer: The views expressed are mine. The objective of research is to enhance our knowledge with breadth and depth and so does not have a ‘one size fits all’ approach. You may find variations of the structure (framework), but the main coverage of a research paper mostly includes what is discussed in section 1 through section.