Business forecasting during the pandemic.

Area of Science:

• Economics
• Econometrics
• Business Analytics

Background:

• The COVID-19 pandemic introduced unprecedented economic uncertainty, challenging established business forecasting methods.
• Traditional statistical models used by business economists were tested by the pandemic's disruptive impact.

Purpose of the Study:

• To retrospectively evaluate the resilience of common statistical models used in business forecasting during the COVID-19 pandemic shock.
• To identify which forecasting approaches were most robust amidst elevated economic uncertainty.

Main Methods:

• Retrospective evaluation of workhorse statistical models employed by business economists.
• Comparative analysis of projection-based versus iteration-based forecasting approaches.
• Assessment of models incorporating macroeconomic data and alternative high-frequency data.

Main Results:

• Projection-based forecasting methods demonstrated greater resilience to the pandemic shock compared to iteration-based methods.
• The pandemic significantly impacted forecast accuracy across models utilizing macroeconomic data.
• Inclusion of high-frequency data did not consistently enhance forecast performance.

Conclusions:

• Projection-based models offer a more resilient framework for business forecasting in times of severe economic disruption.
• Further research is required to determine the precise impact of high-frequency data on business planning and forecast accuracy post-pandemic.