Hello, if you have any need, please feel free to consult us, this is my wechat: wx91due
Applications of Econometrics
Group Project
Deadline: Monday, 16 March, 8:00
Overview and Instructions
The AoE project consist on replicating and extending a published paper.
• Groups must submit the PDF file generated from the R Markdown document. This PDF must combine the code and the economic analysis. In addition, groups must submit a ZIP compressed folder containing the R Markdown (.Rmd) file and the corresponding generated HTML file.
• The R Markdown file must be written so that, when executed, it reproduces the submitted PDF.
• The economic analysis is limited to 2,000 words. This limit excludes code, tables, and figures.
• The tables should reproduce the main coefficients and patterns reported in the paper. The format and presentation should closely mirror those in the original publication. Minor nu merical differences arising from software implementation or rounding are acceptable.
• Before submission, groups must declare that the project is their own work. No separate form is required; the declaration can be completed directly on Learn.
• The use of generative AI or large language models is not permitted.
Project
The paper “Creativity over time and space: a historical analysis of European cities” by M. Serafinelli and G. Tabellini, published in the Journal of Economic Growth (2022), studies the concentration of creativity across time and space. In particular, the authors focus on the mechanisms that explain the formation and decay of clusters of creativity. They use data on notable individuals born in Europe between the eleventh and the nineteenth centuries with historical city data to document the following findings: (i) The production and attraction of creative talent is associated with city institutions that protected economic and political freedoms and promoted local autonomy. (ii)
Indicators of local economic conditions, such as city size and real wages, do not predict creative clusters. (iii) Famous creatives are spatially concentrated and clustered across disciplines, and their spatial mobility has remained stable over the centuries.
The objective of this exercise is to replicate and extend some of these results in this paper.
1. Replicate Table 5.
(a) Explain the diagonal elements of the matrix and those in the 4th quartile.
(b) Extend this analysis by estimating the transition probabilities across deciles. Compare your findings to those reported in the paper.
2. Replicate the analysis in the upper rows of Table 6, Panels A and B, to study the co agglomeration across arts, education, business, and sciences. Discuss your findings
(a) Can we provide a causal interpretation to the coefficients in this table? If not, propose a thought experiment to capture the causal effect of co-agglomeration.
(b) Re-estimate this analysis using the UK only and discuss whether the trend is similar.
3. Replicate Table 7
(a) Discuss why the coefficients of Log(Population) and L.Log(Population) are very similar across the 4 specifications.
4. Replicate Table 8 columns 1 (Log(1 + Un.Births.)) and 3 (Log(1 + Un.Immig.)).
5. Replicate Table 9.
(a) Discuss what identification assumption is required for a causal interpretation and explain why it may fail in this setting.
(b) Given your answer to point (a), discuss the estimates in Table 9.
6. Replicate Figure 1 and tabulate the estimated coefficients.
(a) Explain what the treated and control groups are in this analysis.
(b) The event-study suggests the absence of pre-trends, can we conclude that the estimates are causal? Explain.
(c) Re-estimate this event-study, given by equation (5) in the paper, with 3 leads and 3 lags.
7. Replicate Table 10.
2(a) Discuss the validity of the instrument Regional Commune.
(b) Re-estimate Table 10 using the lag-value of Regional Commune as an instrument. Does it mitigate any concerns about the validity of this instrument?
(c) Does the evidence suggest that Regional Commune is a weak instrument?
8. Replicate Table 11.
(a) Compare Table 11 estimates with those in Table 10. In particular, do both approaches estimate the same causal effect?
(b) Conduct an exogeneity test on the instrument Regional Commune.
9. Replicate Table 12.
(a) The authors claim “Overall, although these Poisson estimates cannot exploit our instrument for Commune, they confirm the main finding above.” discuss why they cannot
implement an instrumental variables strategy.
(b) Is there any methodology discussed in class that could be used to implement the IV strategy in this setting? If so, explain this approach.
Datasets
To complete this exercise, use the following datasets:
• dataset_creativity_time_space_10years_frequency to conduct the analysis of Table 6 and replicate Table 8.
• dataset_creativity_time_space to replicate Tables 5, 7, 9-12, and Figure 1.