The general error correction model in practice

Lebo, Matthew J.; Kraft, Patrick

doi:10.1177/2053168017713059

Cited by 17 publications

(20 citation statements)

References 36 publications

(123 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our other series have time dependence-that is, the mobility in a city on a given day will be highly correlated with mobility in that city on adjacent days-but the extent to which this is present varies from city to city. While properly diagnosing the properties of each time series for each city would allow us to choose among various modeling options, the relatively short time span in our data makes it diffi cult to rely on tests of stationarity ( Lebo and Kraft 2017 ).…”

Section: Modeling the Policy-mobility Relationshipmentioning

confidence: 99%

Do COVID-19 Policies Affect Mobility Behaviour? Evidence from 75 Canadian and American Cities

Armstrong

Lebo

Lucas

2020

Canadian Public Policy

Self Cite

View full text Add to dashboard Cite

Les auteurs proposent une nouvelle mesure de la vigueur des politiques visant la COVID-19 dans 75 villes canadiennes et américaines et estiment les répercussions de ces politiques sur les profi ls de mobilité dans chaque ville. À l'aide d'un nouveau jeu de données sur cinq indicateurs relatifs aux politiques municipales visant la COVID-19 pour chacune des 75 villes, qu'ils associent à 11 indicateurs relatifs aux politiques provinciales ou étatiques, les auteurs élaborent une mesure quotidienne estimative de la « vigueur » de la combinaison de politiques provinciales ou étatiques et municipales visant la COVID-19 dans chaque ville. Ils estiment ensuite les répercussions de ces politiques sur le comportement de mobilité subséquent au moyen de modèles dynamiques de séries chronologiques. Les résultats qu'obtiennent les auteurs corroborent manifestement les répercussions des politiques sur le comportement de mobilité subséquent, mais les différences qu'ils observent entre les villes canadiennes et américaines sont globalement peu nombreuses. Les auteurs traitent de la signifi cation de ces résultats tant pour la recherche sur les politiques visant la COVID-19 que pour la recherche comparative sur d'autres politiques urbaines dans des environnements où les décisions relatives aux politiques sont prises par plusieurs ordres de gouvernement.

show abstract

Section: Modeling the Policy-mobility Relationshipmentioning

confidence: 99%

Do COVID-19 Policies Affect Mobility Behaviour? Evidence from 75 Canadian and American Cities

Armstrong

Lebo

Lucas

2020

Canadian Public Policy

Self Cite

View full text Add to dashboard Cite

show abstract

“…If there is no cointegration, it is possible to apply vector autoregression models in the first differences. In the presence of cointegration of time series, error correction models (Lebo and Kraft, 2017) and vector autoregressive models at levels (Ren, et al, (2013) are used. Based on the official report of the Regulator, the primary set of indicators characterizing the state of the insurance market of Ukraine has been formed.…”

Section: Marketing and Management Of Innovations 2019 Issuementioning

confidence: 99%

Information Management: the Key Driver of the Economic System’s Development

Bacho

Pukala

Hlibko³

et al. 2019

MMI

View full text Add to dashboard Cite

In the scientific article, an information management model was developed for assessing the influence of regulatory tools on business processes on the example of the insurance market of Ukraine. The main purpose of the research is to determine the role of information management as a key driver of business process development in economic systems (on the example of the Ukrainian insurance market). Systematization literary sources and approaches for solving the problem indicates that the development of innovative tools of public governance for the assessment of the regulatory influence of the state regulatory body on the business processes of insurance market is not completed and requires a more in-depth study. The formation of a system of indicators characterizing the business processes of insurance market, which take into account the influence of the Regulator's tools, is proposed to be implemented in the following sequence: to find out the causal links between the indicators of business processes of insurance market and the measures; to determine the existence of the reaction of business processes’ indicators of insurance market to the Regulator’s measures, taking into account the time gap, which determines the presence of lags in the process of applying the Regulator's measures; to define and formalize the variability of the business processes’ indicators of insurance market under the influence of the Regulator's measures. Then the scale of the state of insurance market was built: firstly, check for normality the distribution of the indicators’ values of these markets; secondly, in compliance with the distribution law to construct the scale of indicators according to the «three-sigma rule»; thirdly, in the case of nonconformance with the normal distribution law and presence of skewness, the method of «three sigma» may be used, but either the arithmetic mean, or the mode of the variation series, or its median, is taken as the reference point. The relevance of the decision of this scientific problem is that a set of models is developed for establishing causal relationships between the indicators characterizing the business processes of the insurance market and government regulation instruments, which are quantitatively determined, are theoretical and practical basis for their possible application for solving universal modeling problems of causal processes of estimating the influence of the taken decisions at any regulation of business processes. To solve the problem the Granger-test, the expanded Dickey-Fuller test, the Phillips-Perron test and the Kwiatkowski-Phillips-Schmidt-Shin test were applied. Time series distribution on stationarity and construction of vector autoregression models are implemented. In order to evaluate the adequacy of the models, a comparison of values of Fisher's, Student's criteria, determination coefficient and adjusted determination coefficient is used. Jarque-Bera criteria are used to test the model for stationarity, stability. The formalization of cause-and-effect relationships through the postulate of the theory of measurements and the construction of scales is performed. The built interval scales are based on the application of the three sigma rule, which made possible to specify the allowable boundary values of the indicators. The obtained results testify the influence of regulatory tools on the business processes of insurance market, which confirms the correctness of the use of this complex of models for the solution of weakly formalized problems of causal nature of universal type. The proposed innovative model can be used as a methodology for developing a set of rational methods for assessing informational influences of management decisions in production systems or marketing research. The results of the research can be used to evaluate the business processes of any market or system. Keywords: business processes, Granger test, informational management, informational model, insurance key driver, services market, regulation tools, scaling, vector autoregression.

show abstract

“…That paper showed that much political science research (including some of our own) incorrectly interpreted De Boef and Keele (2008) to imply that the general error correction model (GECM) was more flexible than it is and we emphasized that using the correct MacKinnon critical values was an important part of cointegration tests when estimating the GECM with nonstationary series (also see Enns et al, 2016). Those insights remain essential points of agreement that Lebo and Kraft (2017) identify. 2.…”

Section: Notesmentioning

confidence: 97%

Moving forward with time series analysis

Enns

Kelly

Masaki

et al. 2017

Research & Politics

View full text Add to dashboard Cite

In a recent Research and Politics article, we showed that for many types of time series data, concerns about spurious relationships can be overcome by following standard procedures associated with cointegration tests and the general error correction model (GECM). Matthew Lebo and Patrick Kraft (LK) incorrectly argue that our recommended approach will lead researchers to identify false (i.e., spurious) relationships. In this article, we show how LK's response is incorrect or misleading in multiple ways. Most importantly, when we correct their simulations, their results reinforce our previous findings, highlighting the utility of the GECM when estimated and interpreted correctly.

show abstract

The general error correction model in practice

Cited by 17 publications

References 36 publications

Do COVID-19 Policies Affect Mobility Behaviour? Evidence from 75 Canadian and American Cities

Do COVID-19 Policies Affect Mobility Behaviour? Evidence from 75 Canadian and American Cities

Information Management: the Key Driver of the Economic System’s Development

Moving forward with time series analysis

Contact Info

Product

Resources

About