Central bank economists and researchers use MATLAB® to prototype, validate, deploy, and share financial and economic models in support of critical policy decisions. With MATLAB, you can:
- Use prebuilt apps and tools to pre-process and visualize data
- Fit, simulate, and forecast complex macroeconomic scenarios using univariate and multivariate econometric models
- Use stochastic techniques to model key financial parameters such as GDP, unemployment, inflation, and the economic impact of exogenous shocks such as political disruption, pandemics, and climate change
- Model and manage stress tests and prudential requirements, and maintain orderly markets using machine learning
- Scale, compute, and store data securely in the cloud and integrate with your organization’s enterprise technology infrastructure
- Get technical support, onsite or in-person training, and expert consultant services
Using MATLAB for Modeling the Economy
- Model complex macroeconomic scenarios to conduct monetary analysis and inform critical monetary policy decisions.
- Use point and click econometric apps for data preprocessing, visualization, and fitting time-series data. Work with univariate and multivariate econometric models in the ARIMA/GARCH family and Bayesian and frequentist vector autoregression models. Share results and generate MATLAB code for repeat use.
- Perform time-based macroeconomic general equilibrium analysis of interactions between economic variables using Dynamic Stochastic General Equilibrium (DSGE) modeling.
- Macroeconomic Modeling and Inflation-Rate Forecasting at the Reserve Bank of New Zealand
- MATLAB Used to Predict Financial Crises in Emerging Markets
- Third-Party DSGE Modeling Tools: Bayesian Estimation, Analysis and Regression (BEAR), Dynare, and the IRIS Macroeconomic Modeling Toolbox
- Model-Based Monetary Policy Analysis and Forecasting
Financial Stability and Economic Modeling
- Model key financial parameters such as the GDP, unemployment, inflation, interest rate curves, volatility, default events, and economic growth.
- Analyze multivariate time-series data with structural breaks and unobserved latent states using Markov Switching models. Create and simulate discrete-time Markov chains and time-invariant or time-varying state-space models.
- Model sector and regional economic impacts from exogenous shocks such as political disruption, pandemics, and the physical and transition risks of climate change.
- Automate highly customized reports for policymakers using MATLAB Report Generator™ or deploying web apps directly with MATLAB Web App Server™.
- Using MATLAB for Macroeconomic Stress Testing (26:29)
- Design of Modern Forecasting and Policy Analysis Systems at Central Banks (51:37)
- Estimating Market-Implied Value with Jump-Diffusion Models
- Simulating the Ramsey-Cass-Koopmans Model Using MATLAB and Simulink
- COVID-19 Research and Development with MATLAB
- Swiss Re Calculates Potential Loss from Natural Disasters
Prudential Regulation, Financial Supervision, and Market Surveillance
- Use MATLAB to model the dynamics and risk management procedures of supervised firms to manage stress tests and prudential requirements.
- Maintain liquid, orderly, and fair markets by analyzing market data, news, and alternative data for suspicious activity.
- Design, calculate, and publish benchmark discount rate yield curves using built-in functions.
- Model and visualize complex integrated financial obligations with graph and network algorithms.
- Explore new ideas using hundreds of built-in MATLAB functions and toolboxes, and collaborate with the wider research community by accessing thousands of functions from the MATLAB Central community.
- Take advantage of the latest advances in machine and deep learning, and natural language processing using point and click apps for training and comparing models.
- Accelerate exploratory programming with Live Editor in MATLAB. Share your work through MATLAB Online™ and MATLAB Drive™.
- Leverage the work of colleagues in different programming languages, such as C, Java, or Python, by calling their code directly from MATLAB.
- Share results by publishing live scripts as HTML, PDF, LaTeX, or Microsoft® Word.
- Create engaging lectures that combine explanatory text, mathematical equations, code, and results.
- Scale complex modeling problems with MATLAB Parallel Server™ and generate CUDA code for speeding up computation on GPUs.
- Source market, economic, and alternative data from leading market data providers such as Bloomberg, Refinitiv™, FactSet, and FRED®.
- Access widely used data-sets from statistical agencies using connectors like SDMX, and pull data directly from the web and access any service that exposes a REST API.
- Work with databases and data hubs graphically using the Database Explorer app.
- Access big data storage such as AWS® S3 and Azure® Blob, and use MapReduce solutions such as Spark™, Hadoop®, and DataBricks. Deploy analytics solutions in a microservices architecture managed centrally with MATLAB Production Server™, deploy web applications with MATLAB Web App Server, and scale computations with MATLAB Parallel Server.
Enterprise Technology Infrastructure
- Provide robust, secure, and responsive services by deploying MATLAB solutions in the cloud or datacenter.
- Deploy MATLAB in the cloud or datacenter for both interactive and production use to bring ‘compute to the data’ in a scalable, secure, and flexible way while minimizing the total-cost-of-ownership and management overhead for IT departments.
- Provision and manage flexible compute resources using containerization technologies such as Docker; reference architectures for cloud platforms; and take advantage of specialized hardware such as GPU acceleration.
- Integrate MATLAB models with Enterprise Dashboards Business Information Systems such as Tableau, SpotFire, Power BI, and Qlik.
- Access full technical support, receive online or in-person training, and engage expert consultants.