Incorporating Uncertainties into Economic Forecasts: an Application to Forecasting Economic Activity in Croatia Dario Rukelj Ministry of Finance of the.

Slides:



Advertisements
Similar presentations
Structural reliability analysis with probability- boxes Hao Zhang School of Civil Engineering, University of Sydney, NSW 2006, Australia Michael Beer Institute.
Advertisements

Brief Review –Forecasting for 3 weeks –Simulation Motivation for building simulation models Steps for developing simulation models Stochastic variables.
Hypothesis testing and confidence intervals by resampling by J. Kárász.
E(X 2 ) = Var (X) = E(X 2 ) – [E(X)] 2 E(X) = The Mean and Variance of a Continuous Random Variable In order to calculate the mean or expected value of.
SE503 Advanced Project Management Dr. Ahmed Sameh, Ph.D. Professor, CS & IS Project Uncertainty Management.
CHAPTER 21 Inferential Statistical Analysis. Understanding probability The idea of probability is central to inferential statistics. It means the chance.
An Introduction to Stochastic Reserve Analysis Gerald Kirschner, FCAS, MAAA Deloitte Consulting Casualty Loss Reserve Seminar September 2004.
Resampling techniques Why resampling? Jacknife Cross-validation Bootstrap Examples of application of bootstrap.
ENBIS/1 © Chris Hicks University of Newcastle upon Tyne An analysis of the use of the Beta distribution for planning large complex projects Chris Hicks,
Descriptive statistics Experiment  Data  Sample Statistics Experiment  Data  Sample Statistics Sample mean Sample mean Sample variance Sample variance.
Bootstrap in Finance Esther Ruiz and Maria Rosa Nieto (A. Rodríguez, J. Romo and L. Pascual) Department of Statistics UNIVERSIDAD CARLOS III DE MADRID.
Brief Review –Forecasting for 3 weeks –Simulation Motivation for building simulation models Steps for developing simulation models Stochastic variables.
Introduction to Simulation. What is simulation? A simulation is the imitation of the operation of a real-world system over time. It involves the generation.
2008 Chingchun 1 Bootstrap Chingchun Huang ( 黃敬群 ) Vision Lab, NCTU.
Market Risk VaR: Historical Simulation Approach
Risk Premium Puzzle in Real Estate: Are real estate investors overly risk averse? James D. Shilling DePaul University Tien Foo Sing National University.
Overview of Robust Methods Analysis Jinxia Ma November 7, 2013.
Principles of the Global Positioning System Lecture 10 Prof. Thomas Herring Room A;
Bootstrap spatobotp ttaoospbr Hesterberger & Moore, chapter 16 1.
P á l Rakonczai, L á szl ó Varga, Andr á s Zempl é ni Copula fitting to time-dependent data, with applications to wind speed maxima Eötvös Loránd University.
1 Assessment of Imprecise Reliability Using Efficient Probabilistic Reanalysis Farizal Efstratios Nikolaidis SAE 2007 World Congress.
Transition Matrix Theory and Loss Development John B. Mahon CARe Meeting June 6, 2005 Instrat.
Advanced Risk Management I Lecture 6 Non-linear portfolios.
Comments on “ Incorporating Uncertainties into Economic Forecasts: an Application to Forecasting Economic Activity in Croatia ” Dario Rukelj and Barbara.
Applications of bootstrap method to finance Chin-Ping King.
The Effects of Ranging Noise on Multihop Localization: An Empirical Study from UC Berkeley Abon.
Presentation Ling Zhang Date: Framework of the method 1 Using Distribution Fitting for Assumptions 2 monte – carlo simulation 3 compare different.
Chapter 14 Monte Carlo Simulation Introduction Find several parameters Parameter follow the specific probability distribution Generate parameter.
Probabilistic Mechanism Analysis. Outline Uncertainty in mechanisms Why consider uncertainty Basics of uncertainty Probabilistic mechanism analysis Examples.
Case Study - Relative Risk and Odds Ratio John Snow’s Cholera Investigations.
Theory of Probability Statistics for Business and Economics.
EE325 Introductory Econometrics1 Welcome to EE325 Introductory Econometrics Introduction Why study Econometrics? What is Econometrics? Methodology of Econometrics.
Two Approaches to Calculating Correlated Reserve Indications Across Multiple Lines of Business Gerald Kirschner Classic Solutions Casualty Loss Reserve.
Advanced Higher Statistics Data Analysis and Modelling Hypothesis Testing Statistical Inference AH.
Debt Management, Fiscal Vulnerability and Fiscal Solvency: The Recent Mexican Experience XXII Meeting of the Latin American Network of Central Banks and.
Cointegrating VAR Models and Probability Forecasting: Applied to a Small Open Economy Gustavo Sánchez April 2009.
Statistics - methodology for collecting, analyzing, interpreting and drawing conclusions from collected data Anastasia Kadina GM presentation 6/15/2015.
Analysing shock transmission in a data-rich environment: A large BVAR for New Zealand Chris Bloor and Troy Matheson Reserve Bank of New Zealand Discussion.
Lecture 2 Basics of probability in statistical simulation and stochastic programming Leonidas Sakalauskas Institute of Mathematics and Informatics Vilnius,
Estimating  0 Estimating the proportion of true null hypotheses with the method of moments By Jose M Muino.
Asmah Mohd Jaapar  Introduction  Integrating Market, Credit and Operational Risk  Approximation for Integrated VAR  Integrated VAR Analysis:
Limits to Statistical Theory Bootstrap analysis ESM April 2006.
Probability = Relative Frequency. Typical Distribution for a Discrete Variable.
Reading Report: A unified approach for assessing agreement for continuous and categorical data Yingdong Feng.
Statistical Data Analysis 2011/2012 M. de Gunst Lecture 3.
FORECASTING METHODS OF NON- STATIONARY STOCHASTIC PROCESSES THAT USE EXTERNAL CRITERIA Igor V. Kononenko, Anton N. Repin National Technical University.
1 Chapter 8: Model Inference and Averaging Presented by Hui Fang.
Lecture 3 Types of Probability Distributions Dr Peter Wheale.
Matrix Models for Population Management & Conservation March 2014 Lecture 10 Uncertainty, Process Variance, and Retrospective Perturbation Analysis.
Gil McVean, Department of Statistics Thursday February 12 th 2009 Monte Carlo simulation.
Bootstrapping James G. Anderson, Ph.D. Purdue University.
Estimating standard error using bootstrap
Standard Errors Beside reporting a value of a point estimate we should consider some indication of its precision. For this we usually quote standard error.
Advanced Higher Statistics
Statistical Analysis Urmia University
Ch8 Time Series Modeling
Chapter 7 Sampling Distributions.
CHAPTER 16 ECONOMIC FORECASTING Damodar Gujarati
Estimates of Bias & The Jackknife
Introductory Econometrics
Chapter 7 Sampling Distributions.
Bootstrap - Example Suppose we have an estimator of a parameter and we want to express its accuracy by its standard error but its sampling distribution.
Lecture 2 – Monte Carlo method in finance
Stochastic Hydrology Hydrological Frequency Analysis (I) Fundamentals of HFA Prof. Ke-Sheng Cheng Department of Bioenvironmental Systems Engineering.
Chapter 7 Sampling Distributions.
Chapter 7 Sampling Distributions.
Distributions and Densities
Chapter 7 Sampling Distributions.
Model uncertainty and monetary policy at the Riksbank
Professor Ke-Sheng Cheng
Presentation transcript:

Incorporating Uncertainties into Economic Forecasts: an Application to Forecasting Economic Activity in Croatia Dario Rukelj Ministry of Finance of the Republic of Croatia Barbara Ulloa Central Bank of Chile Young Economists’ Seminar (YES) Dubrovnik Economic Conference June 23, 2010

Uncertainty is the only certainty there is, and knowing how to live with insecurity is the only security... John Allen Paulos

1. INTRODUCTION 2. METHODOLOGY 3. RESULTS 4. CONCLUSION

MOTIVATION FORECASTS OF THE EUROZONE REAL GDP GROWTH

DEALING WITH UNCERTAINTY  Point forecasts  Mode of distribution  Interval forecasts  Consists of an upper and a lower limit  Density forecasts  The whole probability distribution of the forecasts

1. INTRODUCTION 2. METHODOLOGY 3. RESULTS 4. CONCLUSION

STOCHASTIC SIMULATION APPROACH *  Data generating process assumed to be VAR model, estimated in the finite sample:  Forecasts incorporating future uncertainties:  Forecasts incorporating future and parameter uncertainties:  Simulate s in sample values of y  For each of these estimated models, r replications of the forecasts are calculated * Garrat, Pessaran and Shin (2003 and 2006)

SIMULATED SHOCKS  Parametric approach :  Non-parametric approach:  random draws with replacements from the in sample residuals  Unbalanced risks:

CALCULATION  Future uncertainty  Obtain the set of simulated shocks  Generate the forecasts using the simulated shocks  Sort the forecasted values of the variable of interest  Determine probability bands by the deciles  Future and parameter uncertainty  Using initial values for the number of lags determined by the order of the VAR, calculate forecasts ahead using estimated parameters of the initial model, as well as applying a shock to each observation in each period  Re estimate the models with each set of time series obtained in this way  Based on these models forecasts are made like under only future uncertainty

PRESENTATION Probability density 90% 80% 70% 50% 60% Probability Distribution Fanchart

EVALUATION In Sample Fancharts Probability Integral Transform

KOLMOGOROV – SMIRNOV TEST  Kolmogorov – Smirnov test can be used for comparing two distributions  Comparing Probability Integral Transform of outturns with uniform distribution  Let F(a) be the cumulative distribution function of uniform distribution  Cumulative distribution function of empirical distribution is given by: where t is the number of observations of variable b such that  If variable b comes from uniform distribution then D should be small

1. INTRODUCTION 2. METHODOLOGY 3. RESULTS 4. CONCLUSION

Reduced form VECM from Rukelj (2010) considered: where x t is vector of endogenous variables (m, g and y). Rewritten in a VAR form: PORTMANTEAU TEST (H0:Rh=(r1,...,rh)=0) Tested order:10 Adjusted test statistic p-Value:0.151 JARQUE-BERA TEST VariableTest Statisticp-ValueSkewnessKurtosis u u u BENCHMARK MODEL

FUTURE UNCERTAINTY Fanchart – Parametric Approach PIT – Parametric Approach

FUTURE UNCERTAINTY Fanchart – Non Parametric Approach PIT – Non Parametric Approach

FUTURE UNCERTAINTY Fanchart – Skewed Distribution PIT – Skewed Distribution

FUTURE AND PARAMETER UNCERTAINTY Fanchart – Parametric Approach PIT – Parametric Approach

FUTURE AND PARAMETER UNCERTAINTY Fanchart – Non Parametric Approach PIT – Non Parametric Approach

KOLMOGOROV – SMIRNOV TEST RESULTS

1. INTRODUCTION 2. METHODOLOGY 3. RESULTS 4. CONCLUSION

FORECASTING WITH UNCERTAINTY Probability Forecasts for the Real GDP Growth

CONCLUSION  In this paper we have shown how to calculate, present and evaluate density forecasts by stochastic simulation approach  An application of this methodological framework to the chosen benchmark model showed that:  parametric and non-parametric approach yielded similar results  incorporating parameter uncertainty results in a much wider probability bands of the forecasts  evaluation of the density forecasts indicate a better performance when only future, without parameter uncertainties are considered  Future research in this topic should incorporate model uncertainty and additional goodness of fit tests

Thank you for your attention!