Write a small research report using your own data • 4,500 words – including in text references – excluding tables and reference list at the end • No appendix • Analyse the data using Gretl Structure • Introduction: – Relevance, cite other studies – Research question • Methods – Sample and data description – Econometric model specification – Model estimation • Results/Findings – Report estimates (sign, magnitude, signficiance) and results from hypothesis testing (pvalues, Ftest) – Make your own numbered Tables (as few as possible) • Discussion (be critical) – Interpret your finding in terms of you research question – Are my findings believable? If not, why not? – What I found is un/expected? Why? Relate to other studies. – Use common sense in your interpretation of the results – Limitations of the method/data used. • Conclusion – Succinctly states and answers research question Study 1. Review the literature &;amp; develop the theoretical model 2. Specify the model: Select the independent variables &;amp; the functional function. 3. Hypothesize the expected signs of the coefficients. 4. Think whether the CLM assumptions hold and if not what to do about it 5. Estimate &;amp; evaluate the model estimates. Report 1. A brief introduction of the dependent variable and states the research goal and why this research is important. 2. A short review of the relevant literature. 3. An explanation of the model specification. 4. A description of the data &;amp; data sources. 5. A presentation of your final estimated specification. 6. A detailed analysis of the regression results, including any suspected econometric problems 7. A short conclusion that includes any policy implications or recommendations. Writing results • present the results, without elaborate discussion or comment • this does not mean that you do not need any text to describe data presented in tables and figures • All your tables should be numbered and given a title and you need to refer to them in the text. • comment on the significant data presented in the tables and figures Writing discussion • a) consider both sides of an issue, or question before reaching a conclusion • b) consider the results of research and the implications of these. • This is the most complex section in terms of their elements. • Centre around a ‘statement of result’ or an important ‘finding’. • As there is usually more than one result, discussion sections are often structured into a series of discussion cycles. • Use worry questions to be critical of your own research • Discuss if you didn’t resolve any issues pertaining to the fulfillment of the CLM assumptions Computer output • Create a separate document with the computer output giving brief explanations what you did and why. • The computer output is part of the marking. Don’t Do not rephrase your model as a research question – How does Y depend on X1, X2, X3 – Can X1, X2, X3… explain Y • Try many variables in the regression without basing the selection of the candidate variables on economic/social theory. This is known as “data mining”.
crime = total crimes in 1993 / 100,000 population by state (including the District of Columbia)
pov = % of population in state living below the poverty level in 1993
metro = metroplitan population as % of state population
popdens = population per square mile
Source: Statistical Abstract of the United States, 1993, 1994, and 1995