Monitoring Forecasting Performance

Similar documents
A Test for State-Dependent Predictive Ability based on a Markov-Switching Framework

Non-nested model selection. in unstable environments

Improving Equity Premium Forecasts by Incorporating Structural. Break Uncertainty

Time-varying sparsity in dynamic regression models

Improving forecasting performance by window and model averaging

Econ 423 Lecture Notes: Additional Topics in Time Series 1

Out-of-Sample Return Predictability: a Quantile Combination Approach

TECHNICAL WORKING PAPER SERIES APPROXIMATELY NORMAL TEST FOR EQUAL PREDICTIVE ACCURACY IN NESTED MODELS. Todd E. Clark Kenneth D.

Forecasting. Bernt Arne Ødegaard. 16 August 2018

Deep Learning in Asset Pricing

Complete Subset Regressions

Stock Return Predictability Using Dynamic Mixture. Model Averaging

Comparing Predictive Accuracy, Twenty Years Later: On The Use and Abuse of Diebold-Mariano Tests

Elicitability and backtesting

Comparing Possibly Misspeci ed Forecasts

Forecasting the unemployment rate when the forecast loss function is asymmetric. Jing Tian

The Slow Convergence of OLS Estimators of α, β and Portfolio. β and Portfolio Weights under Long Memory Stochastic Volatility

University of Pretoria Department of Economics Working Paper Series

Using all observations when forecasting under structural breaks

Department of Economics, Vanderbilt University While it is known that pseudo-out-of-sample methods are not optimal for

A near optimal test for structural breaks when forecasting under square error loss

General comments Linear vs Non-Linear Univariate vs Multivariate

Robust Backtesting Tests for Value-at-Risk Models

Time-varying parameters: New test tailored to applications in finance and macroeconomics. Russell Davidson and Niels S. Grønborg

Principles of forecasting

Stock Return Prediction with Fully Flexible Models and Coefficients

Forecasting the term structure interest rate of government bond yields

The Empirical Behavior of Out-of-Sample Forecast Comparisons

Comparing Nested Predictive Regression Models with Persistent Predictors

The Comparative Performance of Alternative Out-ofsample Predictability Tests with Non-linear Models

Financial Econometrics Return Predictability

Financial Econometrics

Forecast performance in times of terrorism

Asymptotic Inference about Predictive Accuracy using High Frequency Data

Does modeling a structural break improve forecast accuracy?

Linear models and their mathematical foundations: Simple linear regression

Economic Forecasting with Many Predictors

Forecasting. A lecture on forecasting.

Reality Checks and Nested Forecast Model Comparisons

The regression model with one fixed regressor cont d

Edited by GRAHAM ELLIOTT ALLAN TIMMERMANN

Density Forecast Evaluation in Unstable Environments 1

Multivariate GARCH models.

Forecasting in the presence of recent structural breaks

INFORMATION VALUE ESTIMATOR FOR CREDIT SCORING MODELS

Selecting a Nonlinear Time Series Model using Weighted Tests of Equal Forecast Accuracy

Regression: Ordinary Least Squares

2.5 Forecasting and Impulse Response Functions

Warwick Business School Forecasting System. Summary. Ana Galvao, Anthony Garratt and James Mitchell November, 2014

Two hours. To be supplied by the Examinations Office: Mathematical Formula Tables and Statistical Tables THE UNIVERSITY OF MANCHESTER.

Introduction to Econometrics

Does modeling a structural break improve forecast accuracy?

Miloš Kopa. Decision problems with stochastic dominance constraints

Complex Systems Workshop Lecture III: Behavioral Asset Pricing Model with Heterogeneous Beliefs

Nowcasting GDP directional change with an application to French business survey data

Financial Econometrics and Quantitative Risk Managenent Return Properties

Rolling Window Selection for Out-of-Sample Forecasting with Time-Varying Parameters

Bagging Nonparametric and Semiparametric Forecasts with Constraints

Stock index returns density prediction using GARCH models: Frequentist or Bayesian estimation?

Functional Coefficient Models for Nonstationary Time Series Data

Least angle regression for time series forecasting with many predictors. Sarah Gelper & Christophe Croux Faculty of Business and Economics K.U.

Can a subset of forecasters beat the simple average in the SPF?

Tests of Equal Forecast Accuracy for Overlapping Models

On Generalized Arbitrage Pricing Theory Analysis: Empirical Investigation of the Macroeconomics Modulated Independent State-Space Model

Nowcasting and Short-Term Forecasting of Russia GDP

Research Division Federal Reserve Bank of St. Louis Working Paper Series

A Nonparametric Approach to Identifying a Subset of Forecasters that Outperforms the Simple Average

FORECAST-BASED MODEL SELECTION

Regression Analysis. y t = β 1 x t1 + β 2 x t2 + β k x tk + ϵ t, t = 1,..., T,

Quantile-quantile plots and the method of peaksover-threshold

VARMA versus VAR for Macroeconomic Forecasting

Working Paper Series. Reality Checks and Comparisons of Nested Predictive Models. Todd E. Clark and Michael W. McCracken. Working Paper A

University of Pretoria Department of Economics Working Paper Series

The US Phillips Curve and inflation expectations: A State. Space Markov-Switching explanatory model.

Flexible Inflation Forecast Targeting: Evidence for Canada (and Australia)

Modeling Covariance Risk in Merton s ICAPM

Learning in Real Time: Theory and Empirical Evidence from the Term Structure of Survey Forecasts

Economic Scenario Generation with Regime Switching Models

An economic application of machine learning: Nowcasting Thai exports using global financial market data and time-lag lasso

Are Forecast Updates Progressive?

The Functional Central Limit Theorem and Testing for Time Varying Parameters

Research Brief December 2018

Nested Forecast Model Comparisons: A New Approach to Testing Equal Accuracy

Approximating Fixed-Horizon Forecasts Using Fixed-Event Forecasts

THE LONG-RUN DETERMINANTS OF MONEY DEMAND IN SLOVAKIA MARTIN LUKÁČIK - ADRIANA LUKÁČIKOVÁ - KAROL SZOMOLÁNYI

The Instability of Correlations: Measurement and the Implications for Market Risk

Research Division Federal Reserve Bank of St. Louis Working Paper Series

9) Time series econometrics

Intro VEC and BEKK Example Factor Models Cond Var and Cor Application Ref 4. MGARCH

WORKING PAPER NO DO GDP FORECASTS RESPOND EFFICIENTLY TO CHANGES IN INTEREST RATES?

Forecasting 1: Comparing Forecasting Model

Econ 424 Time Series Concepts

Backtesting Marginal Expected Shortfall and Related Systemic Risk Measures

VAR-based Granger-causality Test in the Presence of Instabilities

Forecasting Macroeconomic Variables Using Diffusion Indexes in Short Samples with Structural Change

Program. The. provide the. coefficientss. (b) References. y Watson. probability (1991), "A. Stock. Arouba, Diebold conditions" based on monthly

FaMIDAS: A Mixed Frequency Factor Model with MIDAS structure

ISyE 6644 Fall 2014 Test 3 Solutions

Model Averaging in Predictive Regressions

Weighted Likelihood Ratio Scores for Evaluating Density Forecasts in Tails

Transcription:

Monitoring Forecasting Performance Identifying when and why return prediction models work Allan Timmermann and Yinchu Zhu University of California, San Diego June 21, 2015

Outline Testing for time-varying forecasting performance Exploiting time-varying forecasting performance A simulation experiment Empirical results

Conditional forecast comparison 1. At time t, we have two predictions ŷ t+1,1 (challenger) and ŷ t+1,2 (benchmark) for the quantity y t+1 2. At time t + 1, we compute the realized relative loss of the two predictions: L t+1 = L (ŷ t+1,1, y t+1 ) L (ŷ t+1,2, y t+1 ). We ask the following questions: Does L t+1 depend on information z t observed at time t? If so, can we exploit this?

Comparison of methodologies Two models ( with parameter β 1 and β 2 with (β 1,, β 2, ) the true parameter value and ˆβt,1, ˆβ ) t,2 the estimated value. L t+1 (β 1, β 2) = L (ŷ t+1,1 (β 1), y t+1) L (ŷ t+1,2 (β 2), y t+1) H (1) 0 : E[ L t+1 ( ˆβ t,1, ˆβ t,2 )] = 0. (Diebold and Mariano (1995)) H (2) 0 : E[ L t+1 (β 1,, β 2, )] = 0. (West (1996), Clark and McCracken (2001), etc) H (3) 0 : E[ L t+1 ( ˆβ t,1, ˆβ t,2 ) z t ] = 0. (Giacomini and White (2006)) H (4) 0 : E[ L t+1 (β 1,, β 2, ) z t ] = 0. We are interested in H (3) 0.

Testing for time-varying forecasting performance H 0 : E ( L t+1 z t ) = 0 almost surely where L t+1 = L t+1 ( ˆβ t,1, ˆβ t,2 ) with ˆβ t,1 and ˆβ t,2 based on a rolling window. Notice that H 0 is equivalent to H 0 : E ( L t+1 h(z t )) = 0 h directing power to chosen directions: Giacomini and White (2006) directing power to all directions: Escanciano (2007)

Giacomini and White (2006) approach We first choose a R d h-valued function h and then test H 0,h : E ( L t+1h (z t )) = 0. [ T 1 J h,t : = T 1 t=1 T 1 ˆΩ h,t = T 1 t=1 L t+1 h (z t ) ] ˆΩ 1 h,t g h,t ( L t+1 ) 2 h (z t ) h (z t ) [ T 1 t=1 L t+1 h (z t ) ] In other words, for a test for H 0 J h,t with χ 2 d h,1 α. with nominal size α, we compare

Escanciano (2007) approach 1. Define R T (u) = T 1/2 T 1 t=1 L t+1w u (z t ) such that E ( L t+1 w u (z t )) = 0 u E ( L t+1 h(z t )) = 0 h e.g. w (z, u) = 1 {z u}. 2. Compute M w,t = R T (u) 2 φ (u) du, where φ ( ) > 0 is some kernel, e.g. pdf of N (0, 1). 3. Simulate M w,t = R T (u) 2 φ (u) du, where R T (u) = T 1/2 T 1 t=1 V t L t+1 w u (z t ) and V t iid with EV t = 0 and EV 2 t = 1. 4. In a test for H 0 of nominal size α, compare M w,t with the 1 α quantile of M w,t.

Exploiting relative forecasting performance If H 0 is reject, we can consider the following simple method: L t+1 = γ 0 + γ 1 z t + ξ t+1 1. At time t, we run the above regression using OLS with a rolling window 2. At time t, our new prediction for y t+1 is ŷ t+1,sw = ŷ t+1,1 1 {ˆγ 0,t + ˆγ 1,t z t > 0}+ŷ t+1,2 1 {ˆγ 0,t + ˆγ 1,t z t 0}

Exploiting relative forecasting performance If E ( L t+1 ) = 0 but E ( L t+1 z t ) 0, then the challenger model is sometimes better but also sometimes worse. The switching rule would be better than always using the challenger model or the benchmark model.

Why should this work: an experiment Why don t we simply include z t in the model? One answer is that it depends on the tradeoff between specification error and estimation error. Consider the following data generating process: y t+1 = α + β st x t + σ st ε t+1 and z t = m st + σ u u t, where x t, ε t, u t are iid N(0, 1) and s t {1, 2} is iid Bernoulli with P (s t = 1) = p.

Why should this work: an experiment We compare the prevailing mean forecast ŷ t+1,0 (benchmark) with the following: ŷ t+1,1 : univariate model y t+1 = α + βx t + ε t+1 using OLS ŷ t+1,2 : bivariate model y t+1 = α + βx t + γz t + ε t+1 using OLS ŷ t+1,3 : true model using MLE ŷ t+1,4 : univariate model y t+1 = α + βx t + ε t+1 with the switching rule described before Let L (i) t+1 = (y t+1 ŷ t+1,0 ) 2 (y t+1 ŷ t+1,i ) 2 and compute, by simulation, E L (i) t+1.

Why should this work: an experiment The parameters in the true data generating process are set to match the real data. All the regressions are done with a rolling window of length 240. We simulate 2.7 million random samples. E L (1) t+1 E L (2) t+1 E L (3) t+1 E L (4) t+1 0.0780 0.0020 0.1028 0.1037

Empirical results: motivation Goyal and Welch (2008): no univariate prediction models seem to outperform out-of-sample the prevailing mean model. Paye and Timmermann (2006), Rapach and Wohar (2006), Goyal and Welch (2008), Rapach, Strauss and Zhou (2010): there are breaks in model parameters; predictability varies with the economic cycle. Henkel, Martin and Nadari (2010), Dangl and Halling (2012), Johannes, Korteweg and Polson (2014): models with regime switching or time-varying coefficients have better performance.

Empirical results: data description We consider the dataset in Goyal and Welch (2008). The goal is to forecast the S&P500 monthly return r t+1. There are 14 predictors, including financial variables: dp (dividende-price ratio), lnv (log realized volatility), etc macro variables: inflation, tbl (t-bill rate) We add more macro variables: UG (unemployment gap), GDP and Cash (firms cash holding).

Empirical results: MSE as loss function Model 1: fit 14 univariate models and use the average of these 14 forecasts as ˆr t+1,1 Model 2: use the forecast of the prevailing mean model as ˆr t+1,2 We look at the p-values of the tests for model instability and the t-stats for E L MSE t+1 and E LMSE t+1,sw, where L MSE t+1 = (r t+1 ˆr t+1,2 ) 2 (r t+1 ˆr t+1,1 ) 2 L MSE t+1,sw = (r t+1 ˆr t+1,2 ) 2 (r t+1 ˆr t+1,sw ) 2

Empirical results: MSE as loss function Z J h,t (p-val) M w,t (p-val) E L MSE t+1 (t-stat) E L MSE t+1,sw (t-stat) X 0.02** 0.23 1.82* 1.97** X 2 0.08* 0.04** 1.20 ln V 0.02** 0.00*** 1.42 UG 0.01** 0.01*** 1.48 GDP 0.04** 0.25 1.41 Cash 0.08* 0.04** 1.28 UG, X 0.02** 0.06* 1.82* UG, X 2 0.03** 0.07* 1.36 UG, ln V 0.02** 0.08* 1.80* UG, GDP 0.03** 0.07* 1.45 UG, Cash 0.03** 0.07* 1.82*

Empirical results: MSE as loss function 20 x 10 3 MSE switching rule of 14 GW var model avg with Z= X 15 10 5 0 5 1970 1975 1980 1985 1990 1995 2000 2005 2010 time Red line is cumulated L MSE t+1,sw ; the blue is cumulated LMSE t+1.

Empirical results: utility as loss function The data and models are exactly the same as before, but we change the loss function. L utility t+1 = U (r t+1, ˆr t+1,1 ) U (r t+1, ˆr t+1,2 ) L utility t+1,sw = U (r t+1, ˆr t+1,sw ) U (r t+1, ˆr t+1,2 ) where U (r, m) = rw (m) γ 2 r 2 w 2 (m), γ measures the risk aversion, and w (m) is the portfolio weight on the risky asset assuming that its conditional mean is m.

Empirical results: utility as loss function Z J h,t (p-val) M w,t (p-val) E L utility t+1 (t-stat) E Lutility t+1,sw (t-stat) X 0.14 0.06* 1.80* 1.85* X 2 0.17 0.01** 1.67* ln V 0.19 0.07* 1.58 UG 0.02** 0.00*** 2.43** GDP 0.19 0.04** 1.79* Cash 0.03** 0.00*** 1.89* UG, X 0.04** 0.01** 2.29** UG, X 2 0.04** 0.02** 2.40** UG, ln V 0.02** 0.02** 2.26** UG, GDP 0.04** 0.02** 2.12** UG, Cash 0.03** 0.02** 2.55**

Empirical results: utility as loss function 1.6 quadratic utility switching rule of 14 GW var model avg with Z= ug 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0.2 1970 1975 1980 1985 1990 1995 2000 2005 2010 time Red line is cumulated L utility t+1,sw ; the blue is cumulated Lutility t+1.