nep-for New Economics Papers
on Forecasting
Issue of 2024‒08‒19
three papers chosen by
Rob J Hyndman, Monash University


  1. Macroeconomic Forecasting with Large Language Models By Andrea Carriero; Davide Pettenuzzo; Shubhranshu Shekhar
  2. Forecasting life expectancy in São Paulo City, Brazil, amidst the COVID-19 pandemic By Maria Laura Miranda; Cássio M. Turra; Ugofilippo Basellini
  3. Calibrated probabilistic forecasts of Finnish TFR 2024–2070 By Ricarda Duerst; Julia Hellstrand; Jonas Schöley; Mikko Myrskylä

  1. By: Andrea Carriero; Davide Pettenuzzo; Shubhranshu Shekhar
    Abstract: This paper presents a comparative analysis evaluating the accuracy of Large Language Models (LLMs) against traditional macro time series forecasting approaches. In recent times, LLMs have surged in popularity for forecasting due to their ability to capture intricate patterns in data and quickly adapt across very different domains. However, their effectiveness in forecasting macroeconomic time series data compared to conventional methods remains an area of interest. To address this, we conduct a rigorous evaluation of LLMs against traditional macro forecasting methods, using as common ground the FRED-MD database. Our findings provide valuable insights into the strengths and limitations of LLMs in forecasting macroeconomic time series, shedding light on their applicability in real-world scenarios
    Date: 2024–06
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2407.00890
  2. By: Maria Laura Miranda (Max Planck Institute for Demographic Research, Rostock, Germany); Cássio M. Turra; Ugofilippo Basellini (Max Planck Institute for Demographic Research, Rostock, Germany)
    Abstract: The COVID-19 pandemic has significantly increased mortality rates, disrupting historical trends and making it challenging to forecast future life expectancy levels. São Paulo, the first city in Brazil to report a COVID-19 case and death saw a decrease of over four years in life expectancy at birth for males and over three years for females between 2019 and 2021. São Paulo has been at the forefront of the demographic transition in the country and experienced a nonlinear mortality decline over the 20th century. The city's historical mortality trajectory and the disruptive effects of COVID-19 have introduced challenges to mortality forecasting. In this study, we used a unique dataset starting from 1920 to forecast life expectancy in São Paulo until 2050 using the Lee-Carter (LC) and Lee-Miller (LM) methods. Mortality rates were obtained from a combination of deaths gathered by the SEADE Foundation (SEADE) and population collected by the Brazilian Institute of Geography and Statistics (IBGE). To mitigate the dependency on the fitting period's choice and better incorporate the effects of the recent mortality shock, we used different baseline periods, using all years from 1920 to 1995 as the starting year of the analysis and six scenarios for post-pandemic mortality levels. Additionally, we used a simulation approach for the time-index parameter to calculate prediction intervals. Based on 73, 200 simulations for each year between 2023 and 2050, we synthesized the resulting life expectancy forecasts into median values and 95% prediction intervals (PI). By 2050, we predict that life expectancy at birth in São Paulo will reach approximately 81.5 years for men and 88.3 years for women. Also, within the 95% PI, we estimated that by 2045, male life expectancy could reach the levels of best-performing countries. Our approach is among the first attempts to forecast mortality in the presence of shocks. Additionally, by evaluating different baseline periods, we advocate for the adoption of more accurate forecasting strategies, particularly in contexts of recent mortality decline. These findings provide valuable resources for policymakers and researchers working to address public health challenges arising from the pandemic and plan for the future well-being of many populations.
    Keywords: Brazil
    JEL: J1 Z0
    Date: 2024
    URL: https://d.repec.org/n?u=RePEc:dem:wpaper:wp-2024-017
  3. By: Ricarda Duerst (Max Planck Institute for Demographic Research, Rostock, Germany); Julia Hellstrand (Max Planck Institute for Demographic Research, Rostock, Germany); Jonas Schöley (Max Planck Institute for Demographic Research, Rostock, Germany); Mikko Myrskylä (Max Planck Institute for Demographic Research, Rostock, Germany)
    JEL: J1 Z0
    Date: 2024
    URL: https://d.repec.org/n?u=RePEc:dem:wpaper:wp-2024-016

This nep-for issue is ©2024 by Rob J Hyndman. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at https://nep.repec.org. For comments please write to the director of NEP, Marco Novarese at <director@nep.repec.org>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.