Time
Start:
End:
Address
zoom Meeting (online only)
Description
October 27, 2020: Adam Gorajek & Joel Bank
Topic: Star Wars at Central Banks
Authors:
Joel Bank (RBA) and Adam Gorajek (RBA, UNSW), on work with co-authors Hamish Fitchett (RBNZ), Andrew Staib (UNSW), and Benjamin Malin (Minneapolis Fed)
Abstract:
We investigate the credibility of central bank research by searching for traces of researcher bias, which is a tendency to use undisclosed analytical procedures that raise measured levels of statistical significance (stars) in artificial ways. To conduct our search we compile a new dataset and borrow two pattern recognition methods from the literature: the p-curve and the z-curve. The p-curve shows no traces of researcher bias but has a high propensity to produce false negatives. The z-curve shows some traces of researcher bias but makes strong assumptions. We examine those assumptions and cast doubt on their merit. At this point, all that is clear is that central banks produce results with patterns different from those in top economic journals, especially near the 5 per cent threshold of statistical significance.
About the speaker:
Adam Gorajek is a PhD candidate at the UNSW and holds bachelor degrees in finance and economics from the University of Adelaide. Aside from doing research, he has worked in various roles at the RBA, most recently as a manager of sections responsible for economic forecasts and monitoring of the domestic banking sector. He has also spent two years as the head of economics at the National Reserve Bank of Tonga and six months at Australia’s banking supervision authority. His research focusses on improving how economists learn about the world, especially methods of economic measurement and econometrics.
Additional information regarding this event will be updated here as it becomes available.
Topic: Star Wars at Central Banks
Authors:
Joel Bank (RBA) and Adam Gorajek (RBA, UNSW), on work with co-authors Hamish Fitchett (RBNZ), Andrew Staib (UNSW), and Benjamin Malin (Minneapolis Fed)
Abstract:
We investigate the credibility of central bank research by searching for traces of researcher bias, which is a tendency to use undisclosed analytical procedures that raise measured levels of statistical significance (stars) in artificial ways. To conduct our search we compile a new dataset and borrow two pattern recognition methods from the literature: the p-curve and the z-curve. The p-curve shows no traces of researcher bias but has a high propensity to produce false negatives. The z-curve shows some traces of researcher bias but makes strong assumptions. We examine those assumptions and cast doubt on their merit. At this point, all that is clear is that central banks produce results with patterns different from those in top economic journals, especially near the 5 per cent threshold of statistical significance.
About the speaker:
Adam Gorajek is a PhD candidate at the UNSW and holds bachelor degrees in finance and economics from the University of Adelaide. Aside from doing research, he has worked in various roles at the RBA, most recently as a manager of sections responsible for economic forecasts and monitoring of the domestic banking sector. He has also spent two years as the head of economics at the National Reserve Bank of Tonga and six months at Australia’s banking supervision authority. His research focusses on improving how economists learn about the world, especially methods of economic measurement and econometrics.
Additional information regarding this event will be updated here as it becomes available.