|
on Forecasting |
| By: | Chow, Hwee Kwan (Singapore Management University); Choy, Keen Meng |
| Abstract: | This paper aims to investigate whether the predictive ability and behaviour of professional forecasters are different during the Covid-19 epidemic as compared with the global financial crisis of 2008 and normal times. To this end, we utilise a survey of professional forecasters in Singapore collated by the central bank to analyse the forecasting record for GDP growth and CPI inflation. We first examine the point forecasts to document the extent of forecast failure in the pandemic crisis and test for behavioural explanations of the possible sources of forecast errors such as leader following and herding behaviour. Using percentile-based summary measures of probability distribution forecasts, we then study how the degree of consensus and subjective uncertainty among forecasters are affected by the heightened economic uncertainty during crises. We found the behaviour of forecasters do not differ much between the two crisis episodes for growth projections despite major differences between the two crises. As for inflation forecasts, our findings suggest forecasters suffer less from forecast inertia when predicting short term one-quarter ahead inflation as compared to longer term one-year and two-year ahead inflation. |
| Keywords: | Survey data; COVID-19; leader following and herding behaviour; disagreement; uncertainty |
| JEL: | D80 E17 |
| Date: | 2022–02–07 |
| URL: | http://d.repec.org/n?u=RePEc:ris:smuesw:2021_010&r= |
| By: | Donato Ceci (Bank of Italy); Andrea Silvestrini (Bank of Italy) |
| Abstract: | This paper compares several methods for constructing weekly nowcasts of recession probabilities in Italy, with a focus on the most recent period of the Covid-19 pandemic. The common thread of these methods is that they use, in different ways, the information content provided by financial market data. In particular, a battery of probit models are estimated after extracting information from a large dataset of more than 130 financial market variables observed at a weekly frequency. The predictive accuracy of these models is explored in a pseudo out-of-sample forecasting exercise. The results demonstrate that nowcasts derived from probit models estimated on a large set of financial variables are, on average, more accurate than standard probit models estimated on a single financial covariate, such as the slope of the yield curve. The proposed approach performs well even compared with probit models estimated on single time series of real economic activity, such as industrial production, or on composite PMI indicators. Overall, the financial indicators used in this paper can be easily updated as soon as new data become available on a weekly basis, thus providing a reliable real-time dating of the Italian business cycle. |
| Keywords: | financial markets, probit models, factor-augmented probit models, model confidence set, penalized likelihood, forecast evaluation |
| JEL: | C22 C25 C53 E32 |
| Date: | 2022–02 |
| URL: | http://d.repec.org/n?u=RePEc:bdi:wptemi:td_1362_22&r= |
| By: | Lang, David Nathan; Wang, Alex; dalal, Nathan; Paepcke, Andreas; Stevens, Mitchell |
| Abstract: | Committing to a major is a fateful step in an undergraduate education, yet the relationship between courses taken early in an academic career and ultimate major selection remains little studied at scale. Using transcript data capturing the academic careers of 26,892 undergraduates enrolled at a private university between 2000 and 2020, we describe enrollment histories using natural-language methods and vector embeddings to forecast terminal major on the basis of course sequences beginning at college entry. We find (I) a student's very first enrolled course predicts major thirty times better than random guessing and more than a third better than majority-class voting, (II) modeling strategies substantially influence forecasting accuracy, and (III) course portfolios varies substantially within majors, raising novel questions what majors mean or signify in relation to undergraduate course histories. |
| Date: | 2021–11–22 |
| URL: | http://d.repec.org/n?u=RePEc:osf:edarxi:u2cwq&r= |