Research Synthesis in Economics and Finance. A course (ECON670) at the Department of Economics and Finance, University of Canterbury, Christchurch, New Zealand, taught in Term 1 of 2025, February and March, when Tomas Havranek was an Erskine Fellow there: ten two-hour lectures and two sessions of student presentations. Tomas Havranek, Charles University, Prague.

Lectures (PDF) Slides as text

From the course description:

Taking stock of research is crucial for both academics and professionals. For example, hedge fund analysts need to understand the key drivers of expected returns. Central bankers calibrate their models based on existing research. Everyone is interested in how an additional year of education affects earnings.

However, individual research findings vary depending on the context. Publication pressures can result in a literature that overstates or understates the true effects. This course delves into meta-analysis, a set of quantitative methods designed for research synthesis. Building on a foundation in econometrics, we will explore modern meta-analysis techniques, emphasizing the correction of publication bias and p-hacking, and discussing their practical applications in economics and finance.

Lectures

1Introduction to Meta-Analysis (19 PDF pages; LaTeX, as text)
2Applied Meta-Analysis Examples (41 PDF pages; LaTeX, as text)
3Literature Search (20 PDF pages; LaTeX, as text)
4Data Collection (21 PDF pages; LaTeX, as text)
5Basic Meta-Analysis Tools (24 PDF pages; LaTeX, as text)
6Student presentations: the replication projects
7Publication Bias (26 PDF pages; LaTeX, as text)
8p-hacking (26 PDF pages; LaTeX, as text)
9Heterogeneity (28 PDF pages; LaTeX, as text)
10Meta-Research (21 PDF pages; LaTeX, as text)
11Limitations of Meta-Analysis (21 PDF pages; LaTeX, as text)
12Student presentations: the applied meta-analyses

The ten decks are the PDFs as taught. The slides as text give every slide's title, text and figures on one page, and latex.zip holds the LaTeX source of each lecture with the figures it uses, ready for pdflatex.

How the course worked

Two lectures of two hours a week, for students with basic statistics and econometrics. Students completed a replication project and an applied meta-analysis project, presented them in the two presentation sessions, and discussed each other's work. The later Stockholm course of August 2025 and Osaka course of March 2026 build on these lectures.

Corrections

Three slides contain errors found after the course. The decks and their LaTeX sources are left as taught; each correction also appears under its slide in the slides as text.

  • FE vs. UWLS: variance. The UWLS variance of the pooled estimate is Σ w_i (θ_i - θ̂)^2 / ((k - 1) Σ w_i); the table leaves out the factor k - 1. This is the formula for independent estimates: with several estimates per study, use standard errors clustered by study.
  • Estimating τ^2 in random-effects model. The REML line is not the restricted likelihood. REML chooses τ^2 ≥ 0 to minimise Σ log(v_i + τ^2) + log Σ W_i + Σ W_i (y_i - θ̂)^2, where v_i are the within-study variances, W_i = 1/(v_i + τ^2) and θ̂ is the weighted mean; software solves this numerically. A negative DL estimate is set to zero.
  • Paxil scandal. The FDA never approved paroxetine for patients under 18: it was approved for adults and promoted for adolescent depression. Study 329 ran from 1994 to 1998 and was published in 2001; the independent reanalysis appeared in 2015.

Readings

Licence

CC BY 4.0, like everything on this site.

A credit line for reuse: Havranek, T. (2025). Research Synthesis in Economics and Finance. Lecture slides, University of Canterbury, Christchurch, February and March 2025. https://meta-analysis.cz/teaching/christchurch-2025/