Search results

We investigate whether smooth robust methods for forecasting can help mitigate pronounced and persistent failure across multiple forecast horizons. We demonstrate that naive predictors are interpretable as local estimators of the long-run relationship with the advantage of adapting quickly after a break, but at a cost of additional forecast error variance. Smoothing over naive estimates helps retain these advantages while reducing the costs, especially for longer forecast horizons. We derive the performance of these predictors after a location shift, and confirm the results using simulations. We apply smooth methods to forecasts of UK productivity and US 10-year Treasury yields and show that they can dramatically reduce persistent forecast failure exhibited by forecasts from macroeconomic models and professional forecasters.

Structural models' inflation forecasts are often inferior to those of naïve devices. This chapter theoretically and empirically assesses this for UK annual and quarterly inflation, using the theoretical framework in Clements and Hendry (1998, 1999). Forecasts from equilibrium-correction mechanisms, built by automatic model selection, are compared to various robust devices. Forecast-error taxonomies for aggregated and time-disaggregated information reveal that the impacts of structural breaks are identical between these, helping to interpret the empirical findings. Forecast failures in structural models are driven by their deterministic terms, confirming location shifts as a pernicious cause thereof, and explaining the success of robust devices.

The authors develop a novel forecast combination approach based on the order statistics of individual predictability from panel data forecasts. To this end, the authors define the notion of forecast depth, which provides a ranking among different forecasts based on their normalized forecast errors during the training period. The forecast combination is in the form of a depth-weighted trimmed mean. The authors derive the limiting distribution of the depth-weighted forecast combination, based on which the authors can readily construct prediction intervals. Using this novel forecast combination, the authors predict the national level of new COVID-19 cases in the United States and compare it with other approaches including the ensemble forecast from the Centers for Disease Control and Prevention (CDC). The authors find that the depth-weighted forecast combination yields more accurate and robust predictions compared with other popular forecast combinations and reports much narrower prediction intervals.

This chapter investigates the performance of different dimension reduction approaches for large vector autoregressions in multi-step ahead forecasts. The authors consider factor augmented VAR models using principal components and partial least squares, random subset regression, random projection, random compression, and estimation via LASSO and Bayesian VAR. The authors compare the accuracy of iterated and direct multi-step point and density forecasts. The comparison is based on macroeconomic and financial variables from the FRED-MD data base. Our findings suggest that random subspace methods and LASSO estimation deliver the most precise forecasts.

In recent work, we have developed a theory of economic forecasting for empirical econometric models when there are structural breaks. This research shows that well-specified models may forecast poorly, whereas it is possible to design forecasting devices more immune to the effects of breaks. In this chapter, we summarise key aspects of that theory, describe the models and data, then provide an empirical illustration of some of these developments when the goal is to generate sequences of inflation forecasts over a long historical period, starting with the model of annual inflation in the UK over 1875–1991 in Hendry (2001a).

Small-scale VARs are widely used in macroeconomics for forecasting US output, prices, and interest rates. However, recent work suggests these models may exhibit instabilities. As such, a variety of estimation or forecasting methods might be used to improve their forecast accuracy. These include using different observation windows for estimation, intercept correction, time-varying parameters, break dating, Bayesian shrinkage, model averaging, etc. This paper compares the effectiveness of such methods in real-time forecasting. We use forecasts from univariate time series models, the Survey of Professional Forecasters, and the Federal Reserve Board's Greenbook as benchmarks.

This study explores the pioneering approach of utilising machine learning (ML) models and integrating social media data for predicting tourist arrivals in Sri Lanka.

Two sets of experiments are performed in this research. First, the predictive accuracy of three ML models, support vector regression (SVR), random forest (RF) and artificial neural network (ANN), is compared against the seasonal autoregressive integrated moving average (SARIMA) model using historical tourist arrivals as features. Subsequently, the impact of incorporating social media data from TripAdvisor and Google Trends as additional features is investigated.

The findings reveal that the ML models generally outperform the SARIMA model, particularly from 2019 to 2021, when several unexpected events occurred in Sri Lanka. When integrating social media data, the RF model performs significantly better during most years, whereas the SVR model does not exhibit significant improvement. Although adding social media data to the ANN model does not yield superior forecasts, it exhibits proficiency in capturing data trends.

The findings offer substantial implications for the industry's growth and resilience, allowing stakeholders to make accurate data-driven decisions to navigate the unpredictable dynamics of Sri Lanka's tourism sector.

This study presents the first exploration of ML models and the integration of social media data for forecasting Sri Lankan tourist arrivals, contributing to the advancement of research in this domain.

The purpose of this paper is to explore whether the out-of-sample model bias plays an important role in predicting volatility.

Under the heterogeneous autoregressive realized volatility (HAR-RV) framework, we analyze the predictive power of out-of-sample model bias for the realized volatility (RV) of the Dow Jones Industrial Average (DJI) and the S&P 500 (SPX) indices from in-sample and out-of-sample perspectives respectively.

The in-sample results reveal that the prediction model including the model bias can obtain bigger R², and the out-of-sample empirical results based on several evaluation methods suggest that the prediction model incorporating model bias can improve forecast accuracy for the RV of the DJI and the SPX indices. That is, model bias can enhance the predictability of original HAR family models.

The author introduce out-of-sample model bias into HAR family models to enhance model capability in predicting realized volatility.