Journal of Forecasting最新文献

Age-specific life-table death counts observed over time are examples of densities. Nonnegativity and summability are constraints that sometimes require modifications of standard linear statistical methods. The centered log-ratio transformation presents a mapping from a constrained to a less constrained space. With a time series of densities, forecasts are more relevant to the recent data than the data from the distant past. We introduce a weighted compositional functional data analysis for modeling and forecasting life-table death counts. Our extension assigns higher weights to more recent data and provides a modeling scheme easily adapted for constraints. We illustrate our method using age-specific Swedish life-table death counts from 1751 to 2020. Compared with their unweighted counterparts, the weighted compositional data analytic method improves short-term point and interval forecast accuracies. The improved forecast accuracy could help actuaries improve the pricing of annuities and setting of reserves.

How do professionals forecast in uncertain times, when the relationships between variables that held in the past may no longer be useful for forecasting the future? For inflation forecasting, we answer this question by measuring survey respondents' adherence to their pre-COVID-19 Phillips curve models during the pandemic. We also ask whether professionals ought to have put their trust in their Phillips curve models over the COVID-19 period. We address these questions allowing for heterogeneity in respondents' forecasts and in their perceptions of the Phillips curve relationship.

Widely used volatility forecasting methods are usually based on low-frequency time series models. Although some of them employ high-frequency observations, these intraday data are often summarized into low-frequency point statistics, for example, daily realized measures, before being incorporated into a forecasting model. This paper contributes to the volatility forecasting literature by instead predicting the next-period intraday volatility curve via a functional time series forecasting approach. Asymptotic theory related to the estimation of latent volatility curves via functional principal analysis is formally established, laying a solid theoretical foundation of the proposed forecasting method. In contrast with nonfunctional methods, the proposed functional approach fully exploits the rich intraday information and hence leads to more accurate volatility forecasts. This is confirmed by extensive comparisons between the proposed method and those widely used nonfunctional methods in both Monte Carlo simulations and an empirical study on a number of stocks and equity indices from the Chinese market.

Market participants who need to trade a significant number of securities within a given period can face high transaction costs. In this paper, we document how improvements in intraday liquidity forecasts can help reduce total transaction costs. We compare various approaches for forecasting intraday transaction costs, including autoregressive and machine learning models, using comprehensive ultra-high-frequency limit order book data for a sample of NYSE stocks from 2002 to 2012. Our results indicate that improved liquidity forecasts can significantly decrease total transaction costs. Simple models capturing seasonality in market liquidity tend to outperform alternative models.

This paper introduces a simple yet effective modification to bootstrap aggregation (bagging) and boosting techniques, aimed at addressing substantial errors arising from parameter estimation, particularly prevalent in macroeconomic and financial forecasting. We propose “egalitarian” bagging and boosting algorithms, where forecasts are derived through an equally weighted combination scheme following variable selection procedures, rather than relying on estimated model parameters. Our empirical work focuses on volatility forecasting, where our approach is applied to a hierarchical model that aggregates a diverse array of volatility components over different time intervals. Significant improvements in predictive accuracy are observed when conventional bagging and boosting approaches are replaced by their “egalitarian” counterparts, across a range of assets and forecast horizons. Notably, these improvements are most pronounced during periods of financial market turmoil, particularly for medium- to long-term predictions. In contrast to boosting, which often yields a sparse model specification, bagging effectively leverages a diverse range of volatility cascades to capture rich information without succumbing to increasing estimation errors. The proposed “egalitarian” algorithm plays a crucial role in facilitating this process, contributing to the superior performance of bagging over other competing approaches.

This paper proposes a novel kernel-based generalized random interval multilayer perceptron (KG-iMLP) method for predicting high-volatility interval-valued returns of crude oil. The KG-iMLP model is constructed by utilizing the $�D_K$ distance based on a kernel function, which outperforms the conventional Euclidean distance. Additionally, the optimal kernel function is estimated using the variance–covariance matrix of the prediction error, contributing to a better understanding of the overall characteristics of interval-valued data. The introduction of the kernel function renders the algorithms used for estimating machine learning parameters ineffective. Therefore, this paper further proposes a backward $�D_K$ distance of accumulative error propagation algorithm to estimate both the kernel function and model parameters, which provides a feasible approach for utilizing kernel function in interval neural networks. In the empirical analysis of weekly and daily returns of WTI crude oil, the superior predictive performance of the proposed method is demonstrated, enabling stable and accurate predictions for both point values and interval values. The model exhibits consistent outstanding performance across different network structures, showcasing the potential of KG-iMLP for crude oil price forecasting.

This paper introduces a novel statistical testing technique known as segmented detrended multifractal fluctuation analysis (SMF-DFA) to analyze the structured scaling properties of financial returns and predict the long-term memory of financial markets. The proposed methodology is applied to assess the efficiency of major cryptocurrencies, expanding upon conventional approaches by incorporating different fluctuation regimes identified through a change-point detection test. A single-factor model is employed to characterize the endogenous factors influencing scaling behavior, leading to the development of a self-explanatory machine learning approach for price forecasting. The proposed method is evaluated using daily data from three major cryptocurrencies spanning from April 2017 to December 2022. The analysis aims to determine whether the digital market has experienced significant changes in recent years and assess whether this has resulted in structured multifractal behavior. The study identifies common periods of local scaling among the three prices, with a noticeable decrease in multifractality observed after 2018. Furthermore, complementary tests on shuffled and surrogate data are conducted to explore the distribution, linear correlation, and nonlinear structure, shedding light on the explanation of structured multifractality to some extent. Additionally, prediction experiments based on neural networks fed with multi-fractionally differentiated data demonstrate the utility of this new self-explanatory algorithm for decision-makers and investors seeking more accurate and interpretable forecasts.

We study the statistical properties of the Bitcoin return series and provide a thorough forecasting exercise. Also, we calibrate state-of-the-art machine learning techniques and compare the results with econometric time series models. The empirical assessment provides evidence that the application of machine learning techniques outperforms econometric benchmarks in terms of forecasting precision for both in- and out-of-sample forecasts. We find that both deep learning architectures as well as complex layers, such as LSTM, do not increase the precision of daily forecasts. Specifically, a simple recurrent neural network describes a sensible choice for forecasting daily return series.

In this study, we use monthly G7 industrial production data, commodity price index data, and commodity currency exchange rate data in a dynamic factor model to examine the global economic factors useful for commodity price prediction. We differentiate between the dynamic factors by specifying a persistent factor and a non-persistent factor, both as a single global factor using all data and as factors for each category of data. The in-sample predictive performances of the three persistent factors together are better than the non-persistent factors and the single global factors. Out-of-sample outcomes based on forecast combinations also support the presence of predictive information in the persistent factors for overall commodity prices and for most sub-categories of commodity price indexes relative to their means. The gains in forecast accuracy are heterogeneous, ranging from 5% to 7% in the 1- to 6-month horizon for overall commodity prices to a high of around 20% for fertilizers in the 12-month horizon in the recent sample. We further show that the information in the persistent factors, especially in the commodity currency exchange rate-based persistent factor, can be integrated with other global measures to further improve the predictive performances of the global measures.

Financial distress occurs when a company cannot meet its financial obligations within a specified timeframe, often owing to prolonged poor operational performance. While studies on financial distress prediction (FDP) use financial ratios (FRs) to forecast distress, they neglect to differentiate long-term (LT) attributes from FRs. To address this gap, our study introduces a novel model that distinguishes between LT and short-term (ST) accounting attributes in FRs. Using data from Taiwanese public companies (1991–2018), our proposed model employs a stacking ensemble classifier to split LT and ST Altman's ratios. This study addresses three key questions: (1) Do models involving split of LT and ST ratios outperform those that combine them? (2) How reliable and robust are these proposed models? (3) What is the proposed model's impact on distress prediction? The results show a significant outperformance of the existing solution, with higher accuracy, lower Type I and Type II errors, and reduced misclassification costs. These models are reliable in handling imbalanced data, proving suitable for real-market investigations. Diverse FR contexts from previous Taiwanese studies validate the distinction between LT and ST features, representing robust performance. This model identifies characteristics of correctly and incorrectly predicted distress in companies, providing nuanced insights into complex distress attributes. This study introduces a pioneering model demonstrating superior predictive accuracy, reliability, and robustness by considering the split between LT and ST accounting attributes. It lays a foundation for future studies to extend and refine the proposed model, offering valuable insights into the complex dynamics of FDP.