Forecasting National Recessions Using State‐Level Data

Owyang, Michael T.; Piger, Jeremy M.; Wall, Howard J.

doi:10.1111/jmcb.12228

Cited by 20 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, a handful of papers have applied methods similar to those used here to identify or forecast recessions. Owyang et al () use a Bayesian model‐averaging methodology to show that state‐level employment data identify turning points in the national economy. Giusto & Piger () use machine‐learning algorithms to evaluate how quickly one can call a recession in real time.…”

Section: Empirical Setupmentioning

confidence: 99%

Predicting Recessions with Leading Indicators: Model Averaging and Selection over the Business Cycle

Berge

2015

Journal of Forecasting

View full text Add to dashboard Cite

Four methods of model selection-equally weighted forecasts, Bayesian model-averaged forecasts, and two models produced by the machine-learning algorithm boosting-are applied to the problem of predicting business cycle turning points with a set of common macroeconomic variables. The methods address a fundamental problem faced by forecasters: the most useful model is simple but makes use of all relevant indicators. The results indicate that successful models of recession condition on different economic indicators at different forecast horizons. Predictors that describe real economic activity provide the clearest signal of recession at very short horizons. In contrast, signals from housing and financial markets produce the best forecasts at longer forecast horizons. A real-time forecast experiment explores the predictability of the 2001 and 2007 recessions.

show abstract

Section: Empirical Setupmentioning

confidence: 99%

Predicting Recessions with Leading Indicators: Model Averaging and Selection over the Business Cycle

Berge

2015

Journal of Forecasting

View full text Add to dashboard Cite

show abstract

“…Many studies now use a variety of methods, including MIDAS (Ghysels, Santa‐Clara and Valkanov 2004; Ghysels, Hill and Motegi 2016), dynamic factor models (Giannone, Reichlin and Small 2008; Aastveit and Trovik 2012; Chernis and Sekkel 2017; Dahlhaus, Guénette and Vasishtha 2017; Bragoli and Fosten 2018), and even an amalgamation of the two—factor MIDAS (Marcellino and Schumacher 2010). These have been complemented by approaches drawing on Bayesian techniques (Owyang, Piger and Wall 2015; Bok et al 2018).…”

Section: Introductionmentioning

confidence: 99%

Place‐Based Policies and Nowcasting

de Silva,

Yanotti,

Sinclair

et al. 2023

Australian Economic Review

View full text Add to dashboard Cite

There is a growing need to gauge local economic activity in real time. Localised economic challenges have been emphasised in the wake of the COVID‐19 pandemic. Real‐time trackers (such as OECD trackers) and other nowcasting applications typically correspond to national or highly aggregated regions. In this discussion paper, we briefly explore how unconventional data might be used to produce nowcasts of local economies. We argue that in the absence of traditional nowcasting metrics, efforts to nowcast local economies need a local perspective, with data capture tailored to address heterogeneity across three domains: (1) resources, (2) people and (3) life.

show abstract

“…Forecasting National-Level Macroeconomic IndicatorsHernández-Murillo andOwyang (2006) and Owyang, Piger and Wall (2015), among others, showed that regionally-disaggregated data can predict national-level data. Here, we consider whether state-level data predicts national-level payroll employment 18. Because national-level data are (typically) not exactly equal to the sum of the state-level data, we level the playing …eld by replacing the BEA's reported national series with the sum of the state-level data-in this case, state-level payroll employment 19.…”

mentioning

confidence: 99%

FRED-SD: A real-time database for state-level data with forecasting applications

Bokun

Young

Kliesen

et al. 2023

International Journal of Forecasting

Self Cite

View full text Add to dashboard Cite

We construct a real-time dataset (FRED-SD) with vintage data for the U.S. states that can be used to forecast both state-level and national-level variables. Our dataset includes approximately 28 variables per state, including labor market, production, and housing variables. We conduct two sets of real-time forecasting exercises. The …rst forecasts state-level labor-market variables using …ve di¤erent models and di¤erent levels of industrially-disaggregated data. The second forecasts a national-level variable exploiting the cross-section of state data. The state-forecasting experiments suggest that large models with industrially-disaggregated data tend to have higher predictive ability for industrially-diversi…ed states. For national-level data, we …nd that forecasting and aggregating state-level data can outperform a random walk but not an autoregression. We compare these real-time data experiments with forecasting experiments using …nal-vintage data and …nd very di¤erent results. Because these …nal-vintage results are obtained with revised data that would not have been available at the time the forecasts would have been made, we conclude that the use of real-time data is essential for drawing proper conclusions about state-level forecasting models.

show abstract

Forecasting National Recessions Using State‐Level Data

Cited by 20 publications

References 42 publications

Predicting Recessions with Leading Indicators: Model Averaging and Selection over the Business Cycle

Predicting Recessions with Leading Indicators: Model Averaging and Selection over the Business Cycle

Place‐Based Policies and Nowcasting

FRED-SD: A real-time database for state-level data with forecasting applications

Contact Info

Product

Resources

About