Le modele ToTEM (pour Terms-of-Trade Economic Model) est le principal modele d’equilibre general dynamique et stochastique qu’utilise la Banque du Canada depuis une quinzaine d’annees. Nous fournissons une description technique de la troisieme mouture du modele, ToTEM III. Deux caracteristiques majeures distinguent ToTEM III des anciennes generations de ToTEM. Premierement, la modelisation de la dette des menages et du marche du logement a ete considerablement affinee. Par exemple, maintenant, le modele : - integre un ensemble de menages emprunteurs qui contribuent a la demande globale de logements - fait une nette distinction entre l’encours et les flux de la dette des menages - presente une combinaison de taux d’interet variables et fixes sur la dette des menages - saisit les effets d’une reglementation du rapport pret-valeur - comprend un nouveau secteur specialise dans la production des biens d’investissement residentiels Deuxiemement, ToTEM III a ete estime a l’aide d’une methode bayesienne de pointe. Ces deux caracteristiques ameliorent considerablement les proprietes empiriques du modele. Nous discutons des nouvelles caracteristiques de ToTEM III et comparons ses reponses a des chocs cles avec celles de ToTEM II. Pour illustrer la facon dont ces nouvelles caracteristiques permettent au personnel de la Banque d’etudier un plus large eventail d’enjeux economiques, nous explorons egalement deux applications en matiere de politiques publiques : - analyse de l’incidence de l’endettement plus eleve des menages sur la sensibilite de la consommation aux taux d’interet - evaluation des effets d’une reglementation du rapport pret-valeur
Jusqu’a tout recemment, peu d’efforts etaient consacres a l’evaluation systematique des differents types de defauts souverains ainsi qu’au calcul de la valeur nominale globale des engagements qu’ils representent. Afin de remedier a cette lacune, la Banque du Canada a developpe une base de donnees exhaustive de defauts souverains, qui se trouve sur son site Web et qui est mise a jour en partenariat avec la Banque d’Angleterre.
We present the structure and features of the International Model for Projecting Activity (IMPACT), a global semi-structural model used to conduct projections and policy analysis at the Bank of Canada. Major blocks of the model are developed based on the rational error correction framework of Kozicki and Tinsley (1999), which allows the model to strike a balance between theoretical structure and empirical performance. IMPACT divides the world economy into six regions: United States, the euro area, Japan, China, oil-importing emerging-market economies and oil-exporting rest of the world. The model features a rich set of cross-border trade and financial linkages that have been shown in the literature to be crucial to explaining global co-movements in business cycles. It is also globally consistent in the sense that both net foreign assets and net exports must be equal to zero at the global level. These cross-region linkages and the global stock-flow consistency allow IMPACT to generate a rigorous and more complete picture of the evolution of the global economy to better inform policy.
The Bank of Canada’s Currency Department has used the Canadian Financial Monitor (CFM) survey since 2009 to track Canadians’ cash usage, payment card ownership and usage, and the adoption of payment innovations. A new online CFM survey was launched in 2018. Because it uses non-probability sampling for data collection, selection bias is very likely. We outline various methods for obtaining survey weights and discuss the associated conditions necessary for these weights to eliminate selection bias. In the end, we obtain calibration weights for the 2018 and 2019 online CFM samples. Our final weights improve upon the default weights provided by the survey company in several ways: (i) we choose the calibration variables based on a fully documented selection procedure that employs machine learning techniques; (ii) we use very up-to-date calibration totals; (iii) for each survey year we obtain two sets of weights, one for the full yearly sample of CFM respondents, the other for the sub-sample of CFM respondents who also filled in the methods-of-payment module of the survey.
This report provides a detailed technical description of a stress test model for investment funds called Ceto.
Risk assessment models are an important component of the Bank’s analytical tool kit for assessing the resilience of the financial system. We describe the Framework for Risk Identification and Assessment (FRIDA), a suite of models developed at the Bank of Canada to quantify the impact of financial stability risks to the broader economy and a range of financial system participants (households, businesses and banks).
This technical report describes sampling, weighting and variance estimation for the Bank of Canada’s 2017 Methods-of-Payment Survey. Under quota sampling, a raking ratio method is implemented to generate weights with both post-stratification and nonparametric nonresponse weight adjustments. In the end, we estimate variances of weighted means and proportions using bootstrap replicate survey weights. Compared with probability sampling, we find that (i) strong assumptions are required to reduce bias when probabilities of selection are unknown, and (ii) multiple weight adjustments for bias reduction inflate variance. Therefore, it is important to focus more on bias than on variance in the context of nonprobability sampling.
We present the daily time series of the outstanding amounts of all Government of Canada marketable debt securities from July 2001 to June 2017.
En 2015, la Banque du Canada a mene une enquete de grande ampleur sur les couts des differents modes de paiement pour les detaillants.
This report provides a detailed technical description of the updated MacroFinancial Risk Assessment Framework (MFRAF), which replaces the version described in Gauthier, Souissi and Liu (2014) as the Bank of Canada’s stress-testing model for banks with a focus on domestic systemically important banks (D-SIBs).
Calibrated weights are created to (a) reduce the nonresponse bias; (b) reduce the coverage error; and (c) make the weighted estimates from the sample consistent with the target population in terms of certain key variables. This technical report details our calibration analysis of singlelocation retailers for the Retailer Survey on the Cost of Payment Methods. We first compare two types of calibration approaches, consisting of (1) traditional calibration, in which calibration is implemented after explicit nonresponse modelling, and (2) nonresponse-embedded calibration, where the nonresponse correction is automatically built in (Sarndal and Lundstrom, 2005). After carefully selecting auxiliary variables, we find minor differences between these two methods. We also examine the effects of trimming, sample size, smoothing and influential units on the calibrated weights, and show that our calibration is robust in view of these considerations.
Nonresponse is a considerable challenge in the Retailer Survey on the Cost of Payment Methods conducted by the Bank of Canada in 2015. There are two types of nonresponse in this survey: unit nonresponse, in which a business does not reply to the entire survey, and item nonresponse, in which a business does not respond to particular questions within the survey. Both types may create a bias when computing statistics such as means and weighted totals for different variables. This technical report analyzes solutions to fix the problem of nonresponse in the survey data. Unit nonresponse is addressed through response probability adjustment, in which response probabilities are modelled using logistic regression (a clustering approach for the unit response probabilities is also considered) and are used in the construction of a set of survey weights. Item nonresponse is addressed through imputation, in which the gradient boosting machine (GBM) and extreme gradient boosting (XGBoost) algorithms are used to predict missing values for variables of interest.
Harmonic task scheduling has many attractive properties, including a utilization bound of 100% under rate-monotonic scheduling and reduced jitter. At the same time, it places a severe constraint on the task period assignment for any application. In this paper, we explore the use of harmonic task scheduling for applications with multiple feedback control tasks. We investigate the properties of harmonic scheduling and give an efficient algorithm to calculate response times for harmonic tasks. We present two algorithms for finding harmonic task periods: one that minimizes the distance from an initial set of non-harmonic periods and one that finds all feasible harmonic periods within a given set of ranges. We apply the algorithms in a control and scheduling co-design procedure, where the goal is to optimize the total performance of a number of control tasks that share a common computing platform. The procedure is evaluated in simulated randomized examples, where it is shown that, in general, harmonic scheduling combined with release offsets gives better control performance than standard, non-harmonic scheduling. (Less)
Cloud computing technology provides the means to share physical resources among multiple users and data center tenants by exposing them as virtual resources. There is a strong industrial drive to use similar technology and concepts to provide timing sensitive services. One such is virtual networking services, so called services chains, which consist of several interconnected virtual network functions. This allows for the capacity to be scaled up and down by adding or removing virtual resources. In this work, we develop a model of a service chain and pose the dynamic allocation of resources as an optimization problem. We design and present a set of strategies to allot virtual network nodes in an optimal fashion subject to latency and buffer constraints.
In this paper, we investigate how liquidity conditions in Canada may affect domestic and/or foreign lending of globally active banks and whether this transmission is influenced by individual bank characteristics. We find that Canadian banks expanded their foreign lending during the recent financial crisis, often through acquisitions of foreign banks. We also find evidence that internal capital markets play a role in the lending activities of globally active Canadian banks during times of heightened liquidity risk.
Le calage est une methode de redressement qui utilise de l’information sur la distribution d’un echantillon et de la population nationale pour determiner la ponderation des participants a une enquete. Le calage vise a ponderer un echantillon afin que sa composition demographique soit representative de la population cible.
The authors describe the key features of a new large-scale Canadian macroeconomic forecasting model developed over the past two years at the Bank of Canada. The new model, called LENS for Large Empirical and Semi-structural model, uses a methodology similar to the Federal Reserve Board’s FRB/US model and the Bank of Canada’s projection model of the U.S. economy (MUSE). LENS is based on a system of estimated reduced-form equations that describe the interactions among key macroeconomic variables. The model strikes a balance between theoretical structure and empirical properties, since most behavioural equations combine forward-looking expectations with adjustment costs. Compared to ToTEM, the Bank’s main model for projection and policy analysis, LENS is more driven by the empirical properties of the data than economic theory and generally provides better out-of-sample forecast performance. In addition, LENS is more disaggregated, thereby allowing the analysis of a broader set of issues related to the economic outlook. These properties will make LENS a useful complement to ToTEM for constructing economic projections at the Bank of Canada.
The Amazon basin is an important player in the global methane cycle. Objectives of this work are to establish a forward and inverse modelling framework on regional scale and to determine the methane budget in the Amazon region. Within the BARCA project (Balanco Atmosferico Regional de Carbono na Amazonia) to airborne measurement campaigns were conducted, one in November 2008 and one in May 2009. The analysis of the methane observations confirms that the Amazon basin is a strong source of methane. The majority of the emissions is found to have biogenic origin, i.e. from wetlands. A comparison of five global methane inversions shows the advantage of using satellite observations in inversion systems. The WRF (Weather Research and Forecasting) Greenhouse Gas model was developed to perform high-resolution simulations of the atmospheric methane distribution in the Amazon region. The newly written code is available within the official WRF-Chem version 3.4 release. Simulations for the two months of the BARCA campaigns with two different wetland models and three different wetland maps were conducted with the WRF Greenhouse Gas model. The comparison to observations indicates that the choice of the wetland map is more important than the choice of the wetland model for a comparison to aircraft observations. Flights with a good representation of the atmospheric transport in the model show a higher correlation between observations and simulations. The two-step regional inversion scheme TM3-STILT was applied to the Amazon region for the year 2009 using observations from the 35 m high TT34 tower. The inversion shows improvements in the representation of the seasonal cycle of the methane emissions in the Amazon basin. However, the determination of the methane budget in the Amazon basin is still highly uncertain.
This report provides a detailed technical description of an updated version of the Terms-of-Trade Economic Model (ToTEM II), which replaced ToTEM (Murchison and Rennison 2006) in June 2011 as the Bank of Canada’s quarterly projection model for Canada.
