Small-sample properties of dimensionality statistics for fitting VAR models to aggregate economic data

Nickelsburg, Gerald

doi:10.1016/0304-4076(85)90118-6

Cited by 30 publications

(21 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mills & Prasad (1992), for example, based on their Monte-Carlo experiments, recommend the Schwarz BIC as the first choice of applied researchers. See also Nickelsburg (1985), Lütkepohl (1985), and Yi & Judge (1988).…”

Section: Discussionmentioning

confidence: 99%

Measuring the strength of cointegration and Granger-causality

Atukeren¹

2005

Applied Economics

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may

show abstract

Section: Discussionmentioning

confidence: 99%

Measuring the strength of cointegration and Granger-causality

Atukeren¹

2005

Applied Economics

View full text Add to dashboard Cite

show abstract

“…The test statistic L is distributed as a chi-square variable with degrees of freedom equal to the number of parametric restrictions being tested, given by n (kj ) where n is the number of endogenous variables in the VAR system.' The final lag order chosen is the largest for which the null hypothesis is rejected for SUCcessive increases at the 5% level of significance (Nickelsburg 1985). To verify the final choice of lag length, the Ljung-Box Q-statistic is used to test for significant residual autocorrelation.…”

Section: Empirical Proceduresmentioning

confidence: 99%

Price Dynamics in International Wheat Markets

Goodwin

Schroeder

1991

Canadian J Agri Economics

View full text Add to dashboard Cite

A vector autoregression model is used to evaluate dynamic relationships among international wheat prices. The effects of freight rates and exchange rates are also considered. Forecast error variance decompositions and impulse response functions are used to investigate price dynamics in six important international wheat markets. The results suggest significant dynamic relationships among prices in different international wheat markets and between the prices and exchange rates and transportation costs. On s'est servi d'un modéle vectoriel à autorégression pour évaluer les relations dynamiques entre les cours du blé sur le marché international. On a également tenu compte de l'inci‐dence des coûts de transport et du taux de change. La décomposition prévue de la variance de l'erreur et les fonctions de réaction ponctuelle ont également permis d'approfondir la dynamique des prix sur six grands marchés internationaux du blé. Les résultats suggérent des relations dynamiques notables entre les prix sur différents marchés internationaux du blé, de meme qu'entre les prix et le taux du change et les coûts de transport.

show abstract

“…Using finite-order vector autoregressive models with various persistent levels and realistic sample sizes, Monte Carlo simulations show that, overall, our criteria outperform conventional competitors. Simulation evidence showed that choosing the correct VAR order versus the best forecasting model is a task that could be achieved in small samples with alternative order selection procedures (see Nickelsburg, 1985;Lütkepohl, 1985Lütkepohl, , 1991Lütkepohl, , 2005. They also produce more efficient multistep (and even stepwise) forecasts since they yield the lowest h-step-ahead forecast mean squared errors for the individual components of the holding pseudo-data to forecast.…”

mentioning

confidence: 99%

“…He also showed that the autoregressive order selected with FPE or AIC is asymptotically efficient, then made the case that SIC is not a good choice since it is not efficient. Simulation evidence showed that choosing the correct VAR order versus the best forecasting model is a task that could be achieved in small samples with alternative order selection procedures (see Nickelsburg, 1985;Lütkepohl, 1985Lütkepohl, , 1991Lütkepohl, , 2005. Based on Hannan and Quinn's (1979) claim that that the concept of consistency loses its significance if one seeks an optimal approximation to an AR.1/ process, Paulsen and Tjøstheim (1985) noted that the fact that AIC is not consistent does not necessarily mean that it is inferior to HQ and SIC.…”

mentioning

confidence: 99%

“…On the other hand, strongly consistent lag-order selection criteria such as the more parsimonious HQ and SIC are rather appropriate for analyzing finite-order VARs. Simulation evidence showed that choosing the correct VAR order versus the best forecasting model is a task that could be achieved in small samples with alternative order selection procedures (see Nickelsburg, 1985;Lütkepohl, 1985Lütkepohl, , 1991Lütkepohl, , 2005. Using finite-order VAR settings with small samples, Lütkepohl (2005) showed that SIC outperforms FPE, AIC and HQ in producing superior short-run forecasts, despite having the tendency to underestimate most the true VAR order.Conversely, Kilian (2001) conjectured that the strong downward model misspecification bias by SIC usually translates into severe bias for VAR statistics or implied high-order dynamics such as impulse response functions (IRFs), especially at higher horizons.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Efficient Multistep Forecast Procedures for Multivariate Time Series

Jouini

2015

Journal of Forecasting

View full text Add to dashboard Cite

Upon the evidence that infinite‐order vector autoregression setting is more realistic in time series models, we propose new model selection procedures for producing efficient multistep forecasts. They consist of order selection criteria involving the sample analog of the asymptotic approximation of the h‐step‐ahead forecast mean squared error matrix, where h is the forecast horizon. These criteria are minimized over a truncation order nT under the assumption that an infinite‐order vector autoregression can be approximated, under suitable conditions, with a sequence of truncated models, where nT is increasing with sample size. Using finite‐order vector autoregressive models with various persistent levels and realistic sample sizes, Monte Carlo simulations show that, overall, our criteria outperform conventional competitors. Specifically, they tend to yield better small‐sample distribution of the lag‐order estimates around the true value, while estimating it with relatively satisfactory probabilities. They also produce more efficient multistep (and even stepwise) forecasts since they yield the lowest h‐step‐ahead forecast mean squared errors for the individual components of the holding pseudo‐data to forecast. Thus estimating the actual autoregressive order as well as the best forecasting model can be achieved with the same selection procedure. Such results stand in sharp contrast to the belief that parsimony is a virtue in itself, and state that the relative accuracy of strongly consistent criteria such as the Schwarz information criterion, as claimed in the literature, is overstated. Our criteria are new tools extending those previously existing in the literature and hence can suitably be used for various practical situations when necessary. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Small-sample properties of dimensionality statistics for fitting VAR models to aggregate economic data

Cited by 30 publications

References 21 publications

Measuring the strength of cointegration and Granger-causality

Measuring the strength of cointegration and Granger-causality

Price Dynamics in International Wheat Markets

Efficient Multistep Forecast Procedures for Multivariate Time Series

Contact Info

Product

Resources

About