Evaluating The Forecast of Densities of Linear and Non-Linear Models: Applications to Output Growth and Unemployment

By Michael P. Clements, Jeremy Smith. 2000

173 citations recorded by the source corpus.

In economics density forecasts are rarely available, and as a result attention has traditionally focused on poit forecasts of the mean and the use of mean square error statistics to represent the loss function. We extend the methods of forecasts density evaluation in Diebold, Gunther and Tay (1997) to compare linear and non-linear model based forecasts of US out put growth and changes in the unemployment rate.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/edf047744d51c7b5afb3f55e7571a889b0ac46aa

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Likelihood-based scoring rules for comparing density forecasts in tails (2011, background): “…applications are density forecasts of output growth and inflation obtained from a variety of sources, including statistical time series models (Clements and Smith, 2000), professional forecasters (Diebold et al., 1999), and central banks and other institutions producing so-called ‘fan charts’…”
Evaluating, Comparing and Combining Density Forecasts Using the KLIC with an Application to the Bank of England and NIESR 'Fan' Charts of Inflation (2005, unknown): “The test statistic LR B is computed as LR B = −2 L(0, 1, 0) − L( µ, σ 2 , ρ) , (4) where L( µ, σ 2 , ρ ) is the value of the exact log-likelihood of a A criticism of Berkowitz's LR test is the maintained assumption of normality; see Clements & Smith (2000) and Bao et al. (2004).”
Assessing probabilistic forecasts of multivariate quantities, with an application to ensemble predictions of surface winds (2008, background): “Clements and Smith (2000, 2002) and De Gooijer (2007) studied a variant for density forecasts in R 2 which is based on the product (or the ratio) of the first (unconditional) and the second (conditional) PIT value.”
Business Cycle Asymmetries (2003, methodology): “output growth by a number of authors, including Tiao and Tsay (1994), Potter (1995), Pesaran and Potter (1997), and Clements and Smith (2000). A statistical analysis of the SETAR model was provided by Tong (1995), and of the STAR model by Luukkonen et al. (1998) and Teräsvirta and Anderson (1992).”
An empirical investigation of the usefulness of ARFIMA models for predicting macroeconomic and financial time series (2006, background): “…testing, but also allow for misspecification amongst competing models; an important Christoffersen (1998), Christoffersen and Diebold (1997), Clements and Smith (2000,2002), Corradi and Swanson (2002), Diebold, Gunther and Tay (1998), Diebold, Hahn and Tay (1999), Harvey, Leybourne and…”
EVALUATING T HE BANK OF ENGLAND D ENSITY FORECASTS OF INFLATION (2004, unknown): “Tests based on the probability integral transform have recently been used to evaluate probability distributions of macroeconomic variables, such as the distributions of expected inflation from the Survey of Professional Forecasters, see Diebold, Tay and Wallis (1999) and Clements (2003), and have be”
Threshold autoregression in economics (2011, background): “Clements and Smith [22] examine the performance of SETAR models for forecast densities of U.”
THE PERFORMANCE OF SETAR MODELS: A REGIME CONDITIONAL EVALUATION OF POINT, INTERVAL AND DENSITY FORECASTS (2004, background): “Previous authors have found that an evaluation based on density forecasts may reveal greater discrimination over the linear models than evaluations based on the first moment (Clements and Smith, 2000, 2001, Boero and Marrocu, 2002a).”
Functional Coefficient Regression Models for Non‐linear Time Series: A Polynomial Spline Approach (2004, methodology): “The approach we used to evaluate the density forecasts is based on Dawid (1984) and Diebold et al. (1998) and was applied in, for example, Clements & Smith (2000).”
Bootstrap Conditional Distribution Tests In the Presence of Dynamic Misspecification (2006, background): “This approach has been shown to be very useful for financial risk management (see e.g. Diebold, Hahn and Tay (1998)), as well as for macroeconomic forecasting (see e.g. Diebold, Tay and Wallis (1998) and Clements and Smith (2000,2002)).”
Comparing Density Forecasts via Weighted Likelihood Ratio Tests (2007, unknown): “For an estimation sample of size m; we then let ̂fm;t be sample mean of the estimation window and ̂gm;t be the conditional mean forecast implied by a Phillips-curve-type model for Xt+1: ̂fm;t = (Xt + :::+Xt m+1)=m (5) ̂gm;t = ̂m;t + ̂m;tXt + ̂m;tut; where u is monthly U.”
Forecasting with DSGE Models (2010, unknown): “…developments (see, e.g., Diebold, Gunther, and Tay, 1998; Christoffersen, 1998; Amisano and Giacomini, 2007) and interesting empirical applications (Diebold, Tay, and Wallis, 1999; Clements and Smith, 2000; Adolfson, Lindé, and Villani, 2007); see also Tay and Wallis (2000) for a survey.”
Pooling of Forecasts (2004, unknown): “7On interval forecasts, see Granger et al. (1989), Chatfield (1993), Christoffersen (1998) and Clements and Taylor (2003), and on density forecasts, Diebold et al. (1998, 1999a, 1999b), Clements and Smith (2000), and the review by Tay and Wallis (2000).”
Economic Forecasting: Some Lessons from Recent Research (2003, background): “…on density forecasting, where the complete probability distribution of possible future outcomes is forecast: see Tay and Wallis (2001) for a survey, Clements and Smith (2000b) for a multi-step application comparing linear and non-linear models, and Diebold, Gunther and Tay (1998) for the role of…”