Forecasting Australia's real house price index: A comparison of time series and machine learning methods

We employ 47 different algorithms to forecast Australian log real house prices and growth rates, and compare their ability to produce accurate out-of-sample predictions. The algorithms, which are specified in both single- and multi-equation frameworks, consist of traditional time series models, machine learning (ML) procedures, and deep learning neural networks. A method is adopted to compute iterated multistep forecasts from nonlinear ML specifications. While the rankings of forecast accuracy depend on the length of the forecast horizon, as well as on the choice of the dependent variable (log price or growth rate), a few generalizations can be made. For one- and two-quarter-ahead forecasts we find a large number of algorithms that outperform the random walk with drift benchmark. We also report several such outperformances at longer horizons of four and eight quarters, although these are not statistically significant at any conventional level. Six of the eight top forecasts (4 horizons × 2 dependent variables) are generated by the same algorithm, namely a linear support vector regressor (SVR). The other two highest ranked forecasts are produced as simple mean forecast combinations. Linear autoregressive moving average and vector autoregression models produce accurate olne-quarter-ahead predictions, while forecasts generated by deep learning nets rank well across medium and long forecast horizons.     

Efficient Multistep Forecast Procedures for Multivariate Time Series

Upon the evidence that infinite‐order vector autoregression setting is more realistic in time series models, we propose new model selection procedures for producing efficient multistep forecasts. They consist of order selection criteria involving the sample analog of the asymptotic approximation of the h‐step‐ahead forecast mean squared error matrix, where h is the forecast horizon. These criteria are minimized over a truncation order n_T under the assumption that an infinite‐order vector autoregression can be approximated, under suitable conditions, with a sequence of truncated models, where n_T is increasing with sample size. Using finite‐order vector autoregressive models with various persistent levels and realistic sample sizes, Monte Carlo simulations show that, overall, our criteria outperform conventional competitors. Specifically, they tend to yield better small‐sample distribution of the lag‐order estimates around the true value, while estimating it with relatively satisfactory probabilities. They also produce more efficient multistep (and even stepwise) forecasts since they yield the lowest h‐step‐ahead forecast mean squared errors for the individual components of the holding pseudo‐data to forecast. Thus estimating the actual autoregressive order as well as the best forecasting model can be achieved with the same selection procedure. Such results stand in sharp contrast to the belief that parsimony is a virtue in itself, and state that the relative accuracy of strongly consistent criteria such as the Schwarz information criterion, as claimed in the literature, is overstated. Our criteria are new tools extending those previously existing in the literature and hence can suitably be used for various practical situations when necessary. Copyright © 2015 John Wiley & Sons, Ltd.     

Multi‐model Forecasts of the West Texas Intermediate Crude Oil Spot Price

We measure the performance of multi‐model inference (MMI) forecasts compared to predictions made from a single model for crude oil prices. We forecast the West Texas Intermediate (WTI) crude oil spot prices using total OECD petroleum inventory levels, surplus production capacity, the Chicago Board Options Exchange Volatility Index and an implementation of a subset autoregression with exogenous variables (SARX). Coefficient and standard error estimates obtained from SARX determined by conditioning on a single ‘best model’ ignore model uncertainty and result in underestimated standard errors and overestimated coefficients. We find that the MMI forecast outperforms a single‐model forecast for both in‐ and out‐of‐sample datasets over a variety of statistical performance measures, and further find that weighting models according to the Bayesian information criterion generally yields superior results both in and out of sample when compared to the Akaike information criterion. Copyright © 2016 John Wiley & Sons, Ltd.     

A hybrid forecasting approach for piece‐wise stationary time series

We consider the problem of forecasting a stationary time series when there is an unknown mean break close to the forecast origin. Based on the intercept‐correction methods suggested by Clements and Hendry (1998) and Bewley (2003), a hybrid approach is introduced, where the break and break point are treated in a Bayesian fashion. The hyperparameters of the priors are determined by maximizing the marginal density of the data. The distributions of the proposed forecasts are derived. Different intercept‐correction methods are compared using simulation experiments. Our hybrid approach compares favorably with both the uncorrected and the intercept‐corrected forecasts. Copyright © 2006 John Wiley & Sons, Ltd.     

Local to unity,long‐horizon forecasting thresholds for model selection in the AR(1)

This article introduces a novel framework for analysing long‐horizon forecasting of the near non‐stationary AR(1) model. Using the local to unity specification of the autoregressive parameter, I derive the asymptotic distributions of long‐horizon forecast errors both for the unrestricted AR(1), estimated using an ordinary least squares (OLS) regression, and for the random walk (RW). I then identify functions, relating local to unity ‘drift’ to forecast horizon, such that OLS and RW forecasts share the same expected square error. OLS forecasts are preferred on one side of these ‘forecasting thresholds’, while RW forecasts are preferred on the other. In addition to explaining the relative performance of forecasts from these two models, these thresholds prove useful in developing model selection criteria that help a forecaster reduce error. Copyright © 2004 John Wiley & Sons, Ltd.     

Spreads versus professional forecasters as predictors of future output change

We examine whether real output forecasts obtained from the Survey of Professional Forecasters efficiently embody information in the term structure spread. To this end, we employ revised data as well as real‐time vintage data, and we also allow for the possible impact of asymmetric loss functions. Assuming quadratic loss, our results suggest that the term structure spread does contain information useful for forecasting not reflected in the survey forecasts, at least over the longest forecast horizon. However, if we allow agents' loss functions to become more negatively skewed with the forecast horizon, then we cannot reject the rationality of the survey forecasts. Copyright © 2009 John Wiley & Sons, Ltd.     

Comparing the DSGE model with the factor model: an out‐of‐sample forecasting experiment

In this paper, we put dynamic stochastic general equilibrium DSGE forecasts in competition with factor forecasts. We focus on these two models since they represent nicely the two opposing forecasting philosophies. The DSGE model on the one hand has a strong theoretical economic background; the factor model on the other hand is mainly data‐driven. We show that incorporating a large information set using factor analysis can indeed improve the short‐horizon predictive ability, as claimed by many researchers. The micro‐founded DSGE model can provide reasonable forecasts for US inflation, especially with growing forecast horizons. To a certain extent, our results are consistent with the prevailing view that simple time series models should be used in short‐horizon forecasting and structural models should be used in long‐horizon forecasting. Our paper compares both state‐of‐the‐art data‐driven and theory‐based modelling in a rigorous manner. Copyright © 2008 John Wiley & Sons, Ltd.     

The use of encompassing tests for forecast combinations

This paper proposes an algorithm that uses forecast encompassing tests for combining forecasts when there are a large number of forecasts that might enter the combination. The algorithm excludes a forecast from the combination if it is encompassed by another forecast. To assess the usefulness of this approach, an extensive empirical analysis is undertaken using a US macroeconomic dataset. The results are encouraging; the algorithm forecasts outperform benchmark model forecasts, in a mean square error (MSE) sense, in a majority of cases. The paper also compares the empirical performance of different approaches to forecast combination, and provides a rule‐of‐thumb cut‐off point for the thick‐modeling approach. Copyright © 2009 John Wiley & Sons, Ltd.     

A comparison of temperature density forecasts from GARCH and atmospheric models

Density forecasts for weather variables are useful for the many industries exposed to weather risk. Weather ensemble predictions are generated from atmospheric models and consist of multiple future scenarios for a weather variable. The distribution of the scenarios can be used as a density forecast, which is needed for pricing weather derivatives. We consider one to 10‐day‐ahead density forecasts provided by temperature ensemble predictions. More specifically, we evaluate forecasts of the mean and quantiles of the density. The mean of the ensemble scenarios is the most accurate forecast for the mean of the density. We use quantile regression to debias the quantiles of the distribution of the ensemble scenarios. The resultant quantile forecasts compare favourably with those from a GARCH model. These results indicate the strong potential for the use of ensemble prediction in temperature density forecasting. Copyright © 2004 John Wiley & Sons, Ltd.     

Forecast Combination for Euro Area Inflation: A Cure in Times of Crisis?

The period of extraordinary volatility in euro area headline inflation starting in 2007 raised the question whether forecast combination methods can be used to hedge against bad forecast performance of single models during such periods and provide more robust forecasts. We investigate this issue for forecasts from a range of short‐term forecasting models. Our analysis shows that there is considerable variation of the relative performance of the different models over time. To take that into account we suggest employing performance‐based forecast combination methods—in particular, one with more weight on the recent forecast performance. We compare such an approach with equal forecast combination that has been found to outperform more sophisticated forecast combination methods in the past, and investigate whether it can improve forecast accuracy over the single best model. The time‐varying weights assign weights to the economic interpretations of the forecast stemming from different models. We also include a number of benchmark models in our analysis. The combination methods are evaluated for HICP headline inflation and HICP excluding food and energy. We investigate how forecast accuracy of the combination methods differs between pre‐crisis times, the period after the global financial crisis and the full evaluation period, including the global financial crisis with its extraordinary volatility in inflation. Overall, we find that forecast combination helps hedge against bad forecast performance and that performance‐based weighting outperforms simple averaging. Copyright © 2017 John Wiley & Sons, Ltd.     

Power transformation models and volatility forecasting

This paper considers the forecast accuracy of a wide range of volatility models, with particular emphasis on the use of power transformations. Where one‐period‐ahead forecasts are considered, the power autoregressive models are ranked first by a range of error metrics. Over longer forecast horizons, however, generalized autoregressive conditional heteroscedasticity models are preferred. A value‐at‐risk‐based forecast assessment indicates that, while the forecast errors are independent, they are not independent and identically distributed, although this latter result is sensitive to the choice of forecast horizon. Our results are robust across a number of different asset markets. Copyright © 2008 John Wiley & Sons, Ltd.     

Bootstrap prediction bands for forecast paths from vector autoregressive models

The problem of forecasting from vector autoregressive models has attracted considerable attention in the literature. The most popular non‐Bayesian approaches use either asymptotic approximations or bootstrapping to evaluate the uncertainty associated with the forecast. The practice in the empirical literature has been to assess the uncertainty of multi‐step forecasts by connecting the intervals constructed for individual forecast periods. This paper proposes a bootstrap method of constructing prediction bands for forecast paths. The bands are constructed from forecast paths obtained in bootstrap replications using an optimization procedure to find the envelope of the most concentrated paths. From extensive Monte Carlo study, it is found that the proposed method provides more accurate assessment of predictive uncertainty from the vector autoregressive model than its competitors. Copyright © 2010 John Wiley & Sons, Ltd.     

A Quantile Regression Approach to Equity Premium Prediction

We propose a quantile regression approach to equity premium forecasting. Robust point forecasts are generated from a set of quantile forecasts using both fixed and time‐varying weighting schemes, thereby exploiting the entire distributional information associated with each predictor. Further gains are achieved by incorporating the forecast combination methodology into our quantile regression setting. Our approach using a time‐varying weighting scheme delivers statistically and economically significant out‐of‐sample forecasts relative to both the historical average benchmark and the combined predictive mean regression modeling approach. Copyright © 2014 John Wiley & Sons, Ltd.     

Forecasting Monetary Policy Decisions in Australia: A Forecast Combinations Approach

This paper applies a triple‐choice ordered probit model, corrected for nonstationarity to forecast monetary decisions of the Reserve Bank of Australia. The forecast models incorporate a mix of monthly and quarterly macroeconomic time series. Forecast combination is used as an alternative to one multivariate model to improve accuracy of out‐of‐sample forecasts. This accuracy is evaluated with scoring functions, which are also used to construct adaptive weights for combining probability forecasts. This paper finds that combined forecasts outperform multivariable models. These results are robust to different sample sizes and estimation windows. Copyright © 2011 John Wiley & Sons, Ltd.     

Moving average threshold heterogeneous autoregressive (MAT-HAR) models

We propose moving average threshold heterogeneous autoregressive (MAT-HAR) models as a novel combination of heterogeneous autoregression (HAR) and threshold autoregression (TAR). The MAT-HAR has multiple groups of lags of a target series, and a threshold term can appear in each group. The threshold is a moving average of lagged target series, which guarantees time-varying thresholds and simple estimation via least squares. We show via Monte Carlo simulations that the MAT-HAR has sharp in-sample and out-of-sample performance. An empirical application on the industrial production of Japan suggests that significant threshold effects exist, and the MAT-HAR has a higher forecast accuracy than the HAR.     

Combination forecasts of output growth in a seven‐country data set

This paper uses forecast combination methods to forecast output growth in a seven‐country quarterly economic data set covering 1959–1999, with up to 73 predictors per country. Although the forecasts based on individual predictors are unstable over time and across countries, and on average perform worse than an autoregressive benchmark, the combination forecasts often improve upon autoregressive forecasts. Despite the unstable performance of the constituent forecasts, the most successful combination forecasts, like the mean, are the least sensitive to the recent performance of the individual forecasts. While consistent with other evidence on the success of simple combination forecasts, this finding is difficult to explain using the theory of combination forecasting in a stationary environment. Copyright © 2004 John Wiley & Sons, Ltd.     

Point forecasting of intraday volume using Bayesian autoregressive conditional volume models

In this paper, we apply Bayesian inference to model and forecast intraday trading volume, using autoregressive conditional volume (ACV) models, and we evaluate the quality of volume point forecasts. In the empirical application, we focus on the analysis of both in‐ and out‐of‐sample performance of Bayesian ACV models estimated for 2‐minute trading volume data for stocks quoted on the Warsaw Stock Exchange in Poland. We calculate two types of point forecasts, using either expected values or medians of predictive distributions. We conclude that, in general, all considered models generate significantly biased forecasts. We also observe that the considered models significantly outperform such benchmarks as the naïve or rolling means forecasts. Moreover, in terms of root mean squared forecast errors, point predictions obtained within the ACV model with exponential distribution emerge superior compared to those calculated in structures with more general innovation distributions, although in many cases this characteristic turns out to be statistically insignificant. On the other hand, when comparing mean absolute forecast errors, the median forecasts obtained within the ACV models with Burr and generalized gamma distribution are found to be statistically better than other forecasts.     

Forecasting with Specification‐Switching VARs

The specification choices of vector autoregressions (VARs) in forecasting are often not straightforward, as they are complicated by various factors. To deal with model uncertainty and better utilize multiple VARs, this paper adopts the dynamic model averaging/selection (DMA/DMS) algorithm, in which forecasting models are updated and switch over time in a Bayesian manner. In an empirical application to a pool of Bayesian VAR (BVAR) models whose specifications include level and difference, along with differing lag lengths, we demonstrate that specification‐switching VARs are flexible and powerful forecast tools that yield good performance. In particular, they beat the overall best BVAR in most cases and are comparable to or better than the individual best models (for each combination of variable, forecast horizon, and evaluation metrics) for medium‐ and long‐horizon forecasts. We also examine several extensions in which forecast model pools consist of additional individual models in partial differences as well as all level/difference models, and/or time variations in VAR innovations are allowed, and discuss the potential advantages and disadvantages of such specification choices. Copyright © 2016 John Wiley & Sons, Ltd.     

Exchange rate forecasting and the performance of currency portfolios

We examine the potential gains of using exchange rate forecast models and forecast combination methods in the management of currency portfolios for three exchange rates: the euro versus the US dollar, the British pound, and the Japanese yen. We use a battery of econometric specifications to evaluate whether optimal currency portfolios implied by trading strategies based on exchange rate forecasts outperform single currencies and the equally weighted portfolio. We assess the differences in profitability of optimal currency portfolios for different types of investor preferences, two trading strategies, mean squared error‐based composite forecasts, and different forecast horizons. Our results indicate that there are clear benefits of integrating exchange rate forecasts from state‐of‐the‐art econometric models in currency portfolios. These benefits vary across investor preferences and prediction horizons but are rather similar across trading strategies.     

Empirical Bayesian Density Forecasting in Iowa and Shrinkage for the Monte Carlo Era

The track record of a 20‐year history of density forecasts of state tax revenue in Iowa is studied, and potential improvements sought through a search for better‐performing ‘priors’ similar to that conducted three decades ago for point forecasts by Doan, Litterman and Sims (Econometric Reviews, 1984). Comparisons of the point and density forecasts produced under the flat prior are made to those produced by the traditional (mixed estimation) ‘Bayesian VAR’ methods of Doan, Litterman and Sims, as well as to fully Bayesian ‘Minnesota Prior’ forecasts. The actual record and, to a somewhat lesser extent, the record of the alternative procedures studied in pseudo‐real‐time forecasting experiments, share a characteristic: subsequently realized revenues are in the lower tails of the predicted distributions ‘too often’. An alternative empirically based prior is found by working directly on the probability distribution for the vector autoregression parameters—the goal being to discover a better‐performing entropically tilted prior that minimizes out‐of‐sample mean squared error subject to a Kullback–Leibler divergence constraint that the new prior not differ ‘too much’ from the original. We also study the closely related topic of robust prediction appropriate for situations of ambiguity. Robust ‘priors’ are competitive in out‐of‐sample forecasting; despite the freedom afforded the entropically tilted prior, it does not perform better than the simple alternatives. Copyright © 2014 John Wiley & Sons, Ltd.