In this issue we have: Forecasting Intraday Time Series with Multiple Seasonal Cycles Using Parsimonious Seasonal Exponential Smoothing, Forecasting the US unemployment rate, Measuring Forecast Uncertainty by Disagreement, Efficient estimation of forecast uncertainty based on recent forecast errors, and many more.
- Forecasting Intraday Time Series with Multiple Seasonal Cycles Using Parsimonious Seasonal Exponential Smoothing
|
Date: |
2009-10-02 |
|
By: |
James W. Taylor |
|
URL: |
|
|
This paper concerns the forecasting of seasonal intraday time series. An extension of Holt-Winters exponential smoothing has been proposed that smoothes an intraday cycle and an intraweek cycle. A recently proposed exponential smoothing method involves smoothing a different intraday cycle for each distinct type of day of the week. Similar days are allocated identical intraday cycles. A limitation is that the method allows only whole days to be treated as identical. We introduce an exponential smoothing formulation that allows parts of different days of the week to be treated as identical. The result is a method that involves the smoothing and initialisation of fewer terms than the other two exponential smoothing methods. We evaluate forecasting up to a day ahead using two empirical studies. For electricity load data, the new method compares well with a range of alternatives. The second study involves a series of arrivals! at a call centre that is open for a shorter duration at the weekends than on weekdays. By contrast with the previously proposed exponential smoothing methods, our new method can model in a straightforward way this situation, where the number of periods on each day of the week is not the same. |
|
|
Keywords: |
Exponential smoothing; Intraday data; Electricity load; Call centre arrivals. |
|
JEL: |
C22 |
|
Date: |
2009-10-30 |
|
By: |
D'Amuri, Francesco/FD |
|
URL: |
|
|
In this paper we suggest the use of an internet job-search indicator (Google Index, GI) as the best leading indicator to predict the US unemployment rate. We perform a deep out-of-sample comparison of many forecasting models. With respect to the previous literature we concentrate on the monthly series extending the out-of-sample forecast comparison with models that adopt both our preferred leading indicator (GI), the more standard initial claims or combinations of both. Our results show that the GI indeed helps in predicting the US unemployment rate even after controlling for the effects of data snooping. Robustness checks show that models augmented with the GI perform better than traditional ones even in most state-level forecasts and in comparison with the Survey of Professional Forecasters' federal level predictions. |
|
|
Keywords: |
Google econometrics; Forecast comparison; Keyword search; US unemployment; Time series models. |
|
JEL: |
C53 |
|
Date: |
2009 |
|
By: |
Kajal Lahiri |
|
URL: |
|
|
Using a standard decomposition of forecasts errors into common and idiosyncratic shocks, we show that aggregate forecast uncertainty can be expressed as the disagreement among the forecasters plus the perceived variability of future aggregate shocks. Thus, the reliability of disagreement as a proxy for uncertainty will be determined by the stability of the forecasting environment, and the length of the forecast horizon. Using density forecasts from the Survey of Professional Forecasters, we find direct evidence in support of our hypothesis. Our results support the use of GARCH-type models, rather than the ex post squared errors in consensus forecasts, to estimate the ex ante variability of aggregate shocks as a component of aggregate uncertainty. |
|
|
Date: |
2009 |
|
By: |
Knüppel, Malte |
|
URL: |
|
|
Multi-step-ahead forecasts of forecast uncertainty in practice are often based on the horizon-specific sample means of recent squared forecast errors, where the number of available past forecast errors decreases one-to-one with the forecast horizon. In this paper, the efficiency gains from the joint estimation of forecast uncertainty for all horizons in such samples are investigated. Considering optimal forecasts, the efficiency gains can be substantial if the sample is not too large. If forecast uncertainty is estimated by seemingly unrelated regressions, the covariance matrix of the squared forecast errors does not have to be estimated, but simply needs to have a certain structure. In Monte Carlo studies it is found that seemingly unrelated regressions mostly yield estimates which are more efficient than the sample means even if the forecasts are not optimal. Seemingly unrelated regressions are used to address question! s concerning the inflation forecast uncertainty of the Bank of England. |
|
|
Keywords: |
Multi-step-ahead forecasts,forecast error variance,GLS,SUR |
|
JEL: |
C13 |
|
Date: |
2009 |
|
By: |
Kajal Lahiri |
|
URL: |
|
|
Using a Bayesian learning model with heterogeneity across agents, our study aims to identify the relative importance of alternative pathways through which professional forecasters disagree and reach consensus on the term structure of inflation and real GDP forecasts, resulting in different patterns of forecast accuracy. Forecast disagreement arises from two primary sources in our model: differences in the initial prior beliefs, and differences in the interpretation of new public information. Estimated model parameters, together with two separate case studies on (i) the dynamics of forecast disagreement in the aftermath of the 9/11 terrorist attack in the U.S. and (ii) the successful inflation targeting experience in Italy after 1997, firmly establish the importance of these two pathways to expert disagreement, and help to explain the relative forecasting accuracy of these two macroeconomic variables. |
|
|
Date: |
2009 |
|
By: |
Kajal Lahiri |
|
URL: |
|
|
Abstract: We consider how to use information from reported density forecasts from surveys to identify asymmetry in forecasters' loss functions. We show that, for the three common loss functions – Lin-Lin, Linex, and Quad-Quad – we can infer the direction of loss asymmetry by just comparing point forecasts and the central tendency (mean or median) of the underlying density forecasts. If we know the entire distribution of the density forecast, we can calculate the loss function parameters based on the first order condition of forecast optimality. This method is applied to forecasts for annual real output growth and inflation obtained from the Survey of Professional Forecasters (SPF). We find that forecasters treat underprediction of real output growth more dearly than overprediction, reverse is true for inflation. |
|
|
Date: |
2009-05 |
|
By: |
Tomescu-Dumitrescu Cornelia (Constantin Brancusi University of Targu Jiu, Faculty of Economics, Romania) |
|
URL: |
|
|
The forecast of evolution of economic phenomena represent on the most the final objective of econometrics. It withal represent a real attempt of validity elaborate model. Unlike the forecasts based on the study of temporal series which have an recognizable inertial character the forecasts generated by econometric model with simultaneous equations are after to contour the future of ones of important economic variables toward the direct and indirect influences bring the bear on their about exogenous variables. For the relief of the calculus who the realization of the forecasts based on the econometric models its suppose is indicate the use of the specialized informatics programs. One of this is the EViews which is applied because it reduces significant the time who is destined of the econometric analysis and it assure a high accuracy of calculus and of the interpretation of results. |
|
|
Keywords: |
economic phenomena, econometric model, forecasts, EViews program |
|
JEL: |
C1 |
- Signal Extraction and Forecasting of the UK Tourism Income Time Series. A Singular Spectrum Analysis Approach
|
Date: |
2009-09-28 |
|
By: |
Beneki, Christina |
|
URL: |
|
|
We present and apply the Singular Spectrum Analysis (SSA), a relatively new, non-parametric and data-driven method used for signal extraction (trends, seasonal and business cycle components) and forecasting of the UK tourism income. Our results show that SSA outperforms slightly SARIMA and time-varying parameter State Space Models in terms of RMSE, MAE and MAPE forecasting criteria. |
|
|
Keywords: |
Singular Spectrum Analysis; Singular Value Decomposition; Business Cycle Decomposition; Tourism Income; United Kingdom; Signal Extraction; Forecasting |
|
JEL: |
C53 |
|
Date: |
2009-10 |
|
By: |
Michal Andrle |
|
URL: |
|
|
The purpose of the paper is to introduce the new “g3†structural model of the Czech National Bank and illustrate how it is used for forecasting and policy analysis. As from January 2007 the model was regularly used for shadowing official forecasts, and in July 2008 it became the core model of the CNB. In the paper we highlight the most important and unusual features of the model and discuss tools and procedures that help us in forecasting and assessing the economy with the model. The paper is not meant to provide a full derivation of the model or the complete characteristics of its behavior and should not be regarded as model documentation. Rather, the paper demonstrates how the model is used and how it contributes to policy analysis. |
|
|
Keywords: |
DSGE, filtering, forecasting, general equilibrium, monetary policy. |
|
JEL: |
D58 |
|
Date: |
2009-11 |
|
By: |
Anna Matas (GEAP, Dpt. Economia Aplicada. Universitat Autònoma de Barcelona) |
|
URL: |
|
|
Traffic forecasts provide essential input for the appraisal of transport investment projects. However, according to recent empirical evidence, long-term predictions are subject to high levels of uncertainty. This paper quantifies uncertainty in traffic forecasts for the tolled motorway network in Spain. Uncertainty is quantified in the form of a confidence interval for the traffic forecast that includes both model uncertainty and input uncertainty. We apply a stochastic simulation process based on bootstrapping techniques. Furthermore, the paper proposes a new methodology to account for capacity constraints in long-term traffic forecasts. Specifically, we suggest a dynamic model in which the speed of adjustment is related to the ratio between the actual traffic flow and the maximum capacity of the motorway. This methodology is applied to a specific public policy that consists of suppressing the toll on a certain motorway! section before the concession expires. |
|
|
Date: |
2009-10 |
|
By: |
Anthony Garratt |
|
URL: |
|
|
We propose a methodology for producing density forecasts for the output gap in real time using a large number of vector autoregessions in inflation and output gap measures. Density combination utilizes a linear mixture of experts framework to produce potentially non-Gaussian ensemble densities for the unobserved output gap. In our application, we show that data revisions alter substantially our probabilistic assessments of the output gap using a variety of output gap measures derived from univariate detrending filters. The resulting ensemble produces well-calibrated forecast densities for US inflation in real time, in contrast to those from simple univariate autoregressions which ignore the contribution of the output gap. Combining evidence from both linear trends and more flexible univariate detrending filters induces strong multi-modality in the predictive densities for the unobserved output gap. The peaks associated w! ith these two detrending methodologies indicate output gaps of opposite sign for some observations, reflecting the pervasive nature of model uncertainty in our US data. |
|
|
Date: |
2009-10 |
|
By: |
Anthony Garratt |
|
URL: |
|
|
We propose a methodology for producing density forecasts for the output gap in real time using a large number of vector autoregessions in inflation and output gap measures. Density combination utilizes a linear mixture of experts framework to produce potentially non-Gaussian ensemble densities for the unobserved output gap. In our application, we show that data revisions alter substantially our probabilistic assessments of the output gap using a variety of output gap measures derived from univariate detrending filters. The resulting ensemble produces well-calibrated forecast densities for US inflation in real time, in contrast to those from simple univariate autoregressions which ignore the contribution of the output gap. Combining evidence from both linear trends and more flexible univariate detrending filters induces strong multi-modality in the predictive densities for the unobserved output gap. The peaks associated w! ith these two detrending methodologies indicate output gaps of opposite sign for some observations, reflecting the pervasive nature of model uncertainty in our US data. |
|
|
Date: |
2009-10-05 |
|
By: |
Kyungchul Song (Department of Economics, University of Pennsylvania) |
|
URL: |
|
|
One of the approaches to compare forecasts is to test whether the loss from a benchmark prediction is smaller than the others. The test can be embedded into the general problem of testing functional inequalities using a one-sided Kolmogorov-Smirnov functional. This paper shows that such a test generally suffers from unstable power properties, meaning that the asymptotic power against certain local alternatives can be much smaller than the size. This paper proposes a general method to robustify the power properties. This method can also be applied to testing inequalities such as stochastic dominance and moment inequalities. Simulation studies demonstrate that tests based on this paper's approach perform quite well relative to the existing methods. |
|
|
Keywords: |
Inequality Restrictions, Testing Predictive Ability, One-sided Nonparametric Tests, Power Robustification |
|
JEL: |
C12 |
Taken from the NEP-FOR mailing list edited by Rob Hyndman.