Predicting U.S. Business-Cycle Regimes

Birchenhall, Chris; Jessen, Hans; Osborn, Demise R.; Simpson, Paul W.

doi:10.1080/07350015.1999.10524820

Cited by 38 publications

(35 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recession in the beginning of 90s has been especially difficult to forecast, see Fintzen and Stekler (1999). In Stock and Watson (1993), the 1990 recession was missed and Figure 1 in Hamilton and Perez-Quiros (1996) and Figures 1-2 in Birchenhall et al (1999) show that either the indicators are late or they miss that recession completely. In Figure 5 a clear coincident warning is given.…”

Section: The Us Indicatormentioning

confidence: 99%

See 1 more Smart Citation

A classifying procedure for signalling turning points

Koskinen

Öller

2004

Journal of Forecasting

View full text Add to dashboard Cite

Abstract. A Hidden Markov Model (HMM) is used to classify an out of sample observation vector into either of two regimes. This leads to a procedure for making probability forecasts for changes of regimes in a time series, i.e. for turning points.Instead of maximizing a likelihood, the model is estimated with respect to known past regimes. This makes it possible to perform feature extraction and estimation for different forecasting horizons. The inference aspect is emphasized by including a penalty for a wrong decision in the cost function. The method is tested by forecasting turning points in the Swedish and US economies, using leading data. Clear and early turning point signals are obtained, contrasting favourable with earlier HMM studies.Some theoretical arguments for this are given.

show abstract

Section: The Us Indicatormentioning

confidence: 99%

“…3). Note that (11) is the formulation based on information theory suggested in Birchenhall et al (1999).…”

Section: A Dynamic Bayesian Classifiermentioning

confidence: 99%

A classifying procedure for signalling turning points

Koskinen

Öller

2004

Journal of Forecasting

View full text Add to dashboard Cite

show abstract

“…Unless we are confident that we have a model that encompasses the 'true' model, we must treat standard errors and standard t-tests with caution. The procedure adopted here, as in Birchenhall et al (1999), is based on a useful result by Sin and White (1996). If we choose our model so as to maximize an appropriately penalized likelihood function then, asymptotically, we will select the 'best' model even if all models are misspecified 14 .…”

Section: Simulations and An Econometric Model Of Successionmentioning

confidence: 99%

“…The econometric procedures used here are those developed for forecasting business cycle regimes in Birchenhall et al (1999) 13 . There are two key ideas driving this methodology.…”

Section: Simulations and An Econometric Model Of Successionmentioning

confidence: 99%

Structural change in the presence of network externalities: a co-evolutionary model of technological successions

Windrum

Birchenhall

2005

J Evol Econ

View full text Add to dashboard Cite

The paper examines the conditions under which technological successions can occur in the presence of network externalities. A two-stage, multi-agent simulation model is presented in which product designs co-evolve with consumer preferences. It provides a rich framework in which to study the complex phenomenon of quality. Following an initial period, in which old technology firms develop their designs and externalities accrue, a technological shock occurs. New technology firms and new consumer classes enter the market. Data from the simulation model is analysed by identifying a robust econometric model of the probability of succession, given the immediate state of the post-shock market. 4 factors affecting the probability of a succession are identified. First, succession can occur if gains in direct utility from higher quality new technology goods outweigh the network utility of old technology goods. Second, sailing ship effects are possible. Old firms can innovate in order to see off the new entrants. Hence, a better initial (new technology) design does not guarantee succession. Third, a trade-off exists between quality and price. A succession will not occur if cost (price) differentials favour the old technology. Consequently, increasing returns in production enjoyed by established firms are an important barrier to successful entry. The fourth factor is time: the relative length of time old firms have to develop their products, and that which new firms have to develop their products.

show abstract

“…Birchenhall et al (1999) suggest using two rules to convert a predicted probability into a predicted classification. One is the 0.5 rule and the other is the sample rule.…”

Section: Prediction Criteriamentioning

confidence: 99%