|
on Forecasting |
| By: | Gabor Petnehazi; Laith Al Shaggah; Jozsef Gall; Bernadett Aradi |
| Abstract: | This study explores the potential of zero-shot time series forecasting, an innovative approach leveraging pre-trained foundation models, to forecast mortality rates without task-specific fine-tuning. We evaluate two state-of-the-art foundation models, TimesFM and CHRONOS, alongside traditional and machine learning-based methods across three forecasting horizons (5, 10, and 20 years) using data from 50 countries and 111 age groups. In our investigations, zero-shot models showed varying results: while CHRONOS delivered competitive shorter-term forecasts, outperforming traditional methods like ARIMA and the Lee-Carter model, TimesFM consistently underperformed. Fine-tuning CHRONOS on mortality data significantly improved long-term accuracy. A Random Forest model, trained on mortality data, achieved the best overall performance. These findings underscore the potential of zero-shot forecasting while highlighting the need for careful model selection and domain-specific adaptation. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.13521 |
| By: | MINAMI, Koutaroh |
| Abstract: | This study explores the potential of machine learning, Long Short-Term Memory (LSTM), to detect asset price bubbles by analyzing prediction errors. Using monthly data of the Nikkei225 Index, I evaluate the performance of LSTM model in forecasting prices and compare with the GSADF test. I find that LSTM’s prediction accuracy significantly deteriorates during periods associated with asset bubbles, suggesting the presence of structural changes. In particular, the LSTM approach of this paper captures both the emergence and collapse of Japan’s late 1980s bubble separately. In addition, it can also capture structural changes related to policy changes in the 2010s Japan, which are not identified by the GSADF test. These findings suggest that machine learning can be used for not only identifying bubbles but also policy evaluations. |
| Keywords: | Bubbles, Generalized Supremum Augmented Dickey-Fuller test (GSADF), Machine learning, Long Short Term Memory (LSTM) |
| JEL: | G10 G17 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:hit:hcfrwp:g-1-30 |
| By: | Bo Hu; Joon Y. Park; Junhui Qian |
| Abstract: | This paper introduces a novel approach to investigate the dynamics of state distributions, which accommodate both cross-sectional distributions of repeated panels and intra-period distributions of a time series observed at high frequency. In our approach, densities of the state distributions are regarded as functional elements in a Hilbert space, and are assumed to follow a functional autoregressive model. We propose an estimator for the autoregressive operator, establish its consistency, and provide tools and asymptotics to analyze the forecast of state density and the moment dynamics of state distributions. We apply our methodology to study the time series of distributions of the GBP/USD exchange rate intra-month returns and the time series of cross-sectional distributions of the NYSE stocks monthly returns. Finally, we conduct simulations to evaluate the density forecasts based on our model. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.15763 |
| By: | Abdullah Karasan; Ozge Sezgin Alp; Gerhard-Wilhelm Weber |
| Abstract: | In this study, we propose a novel machine-learning-based measure for stock price crash risk, utilizing the minimum covariance determinant methodology. Employing this newly introduced dependent variable, we predict stock price crash risk through cross-sectional regression analysis. The findings confirm that the proposed method effectively captures stock price crash risk, with the model demonstrating strong performance in terms of both statistical significance and economic relevance. Furthermore, leveraging a newly developed firm-specific investor sentiment index, the analysis identifies a positive correlation between stock price crash risk and firm-specific investor sentiment. Specifically, higher levels of sentiment are associated with an increased likelihood of stock price crash risk. This relationship remains robust across different firm sizes and when using the detoned version of the firm-specific investor sentiment index, further validating the reliability of the proposed approach. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.16287 |
| By: | Kerry Xiao; Amy Zang |
| Abstract: | People in the real world often possess vague knowledge of future payoffs, for which quantification is not feasible or desirable. We argue that language, with differing ability to convey vague information, plays an important but less known-role in subjective expectations. Empirically, we find that in their reports, analysts include useful information in linguistic expressions but not numerical forecasts. Specifically, the textual tone of analyst reports has predictive power for forecast errors and subsequent revisions in numerical forecasts, and this relation becomes stronger when analyst's language is vaguer, when uncertainty is higher, and when analysts are busier. Overall, our theory and evidence suggest that some useful information is vaguely known and only communicated through language. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.12269 |
| By: | Escribano, Álvaro; Rodríguez, Juan Andrés |
| Abstract: | The year 2024 marked a critical milestone in global warming, with average global temperatures exceeding pre-industrial levels by 1.55°C— the highest on human historyrecords. Polar ice loss, largely attributed to anthropogenic CO₂ emissions has profound social, economic and financial implications that demand rigorous analysis. This study assesses the impact of atmospheric CO₂ on Arctic and Antarctic sea-ice volume usingnonlinear dynamic econometric models. We extend prior sea-ice forecasting models to allow for regime-switching specifications—Threshold Autoregressive (TAR) and Smooth Transition Regressions (STR) models—to capture the complex, nonlinear, and state-dependent responses of the sea-ice to CO₂ concentration changes. Our main contribution is to provide a flexible, reduced-form alternative to general circulation models (GCMs) for evaluating long-run climate scenarios under various emissions trajectories, including IPCC’s Shared Socioeconomic Pathways (SSPs). Results suggest Arctic sea-ice could disappear by 2060 [2045–2078] under a business-as-usual scenario, while Antarctic loss may extend beyond 2100 [2071–2300]. Importantly, models accounting for threshold effects reveal critical recovery tipping points that simpler linear climate models may overlook. Under an intermediate emissions path like SSP2-4.5, a fast recovery of sea-ice volume remains possible if regime shifts are driven by changes in CO₂ growth rates, with estimated tipping points for reversal occurring around 2033 for the Arctic and 2037 for the Antarctic. In contrast, the outlook is less favorable if regime dynamics are determined by CO₂ concentration levels: no recovery is projected for the Arctic, and the Antarctic recovery tipping point is delayed until 2069. |
| Keywords: | Climate change; Climate econometrics; Sea ice; CO₂; Concentration; General circulation models (GCMs); Shared socioeconomic pathways (SSPs); Tipping points |
| JEL: | C32 C38 C51 C52 C5 Q54 |
| Date: | 2025–07–31 |
| URL: | https://d.repec.org/n?u=RePEc:cte:werepe:47734 |
| By: | Martha Bernate-Valbuena (Department of Economy, Accounting and Finance, University of Monterrey, México); Begoña Gutiérrez (2Department of Accounting and Finance, Universidad de Zaragoza, Spain, School of Economics and Business) |
| Abstract: | This study examines whether earnings management indicators, which highlight unjustified variations in accounting items, can predict business bankruptcy. Using data from 179, 559 Spanish firms, from 2009 to 2014, both traditional financial ratios and earnings management indicators were analyzed. Significant differences between failed and non-failed firms were observed years before bankruptcy. To ensure robustness, a test sample from a future period validated the findings. Logistic regression revealed that certain earnings management indicators, particularly a synthetic index combining multiple indicators, can predict bankruptcy. Such indexes could enhance bankruptcy prediction models, offering valuable insights for assessing financial health and potential risks in businesses. |
| Keywords: | Bankruptcy, financial ratios, earnings management, creative accounting |
| JEL: | G33 G53 |
| Date: | 2025–01 |
| URL: | https://d.repec.org/n?u=RePEc:zar:wpaper:dt2025-01 |
| By: | Takuya Sakaguchi (Graduate School of Economics, Kobe University, JAPAN); Masahiko Shibamoto (Research Institute for Economics and Business Administration and Center for Computational Social Science, Kobe University, JAPAN) |
| Abstract: | Monitoring inflation is an important aspect of policymaking, but understanding the factors that drive inflation remains challenging. In this paper, we construct a cyclically sensitive inflation (CSI) index for Japan and examine its usefulness from several perspectives. Specifically, we investigate (1) the dynamic relationship from financial markets and the real economy to cyclical inflation, (2) how well cyclical inflation can predict future inflation trends, and (3) the effectiveness of the CSI as a real-time indicator of economic slack. Our empirical results show that the CSI can complement headline and core inflation measures and, when used together, help distinguish whether price changes are due to temporary factors or persistent pressures associated with the business cycle. |
| Keywords: | Cyclical inflation; Business cycle; Financial market; Inflation forecast; Japanese economy |
| JEL: | C32 E31 E32 E52 E58 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:kob:dpaper:dp2025-22 |
| By: | Briola, Antonio; Bartolucci, Silvia; Aste, Tomaso |
| Abstract: | We exploit cutting-edge deep learning methodologies to explore the predictability of high-frequency Limit Order Book mid-price changes for a heterogeneous set of stocks traded on the NASDAQ exchange. In so doing, we release ‘LOBFrame’, an open-source code base to efficiently process large-scale Limit Order Book data and quantitatively assess state-of-the-art deep learning models' forecasting capabilities. Our results are twofold. We demonstrate that the stocks' microstructural characteristics influence the efficacy of deep learning methods and that their high forecasting power does not necessarily correspond to actionable trading signals. We argue that traditional machine learning metrics fail to adequately assess the quality of forecasts in the Limit Order Book context. As an alternative, we propose an innovative operational framework that evaluates predictions' practicality by focusing on the probability of accurately forecasting complete transactions. This work offers academics and practitioners an avenue to make informed and robust decisions on the application of deep learning techniques, their scope and limitations, effectively exploiting emergent statistical properties of the Limit Order Book. |
| Keywords: | deep learning; econophysics; high frequency trading; limit order book; market microstructure |
| JEL: | J1 F3 G3 |
| Date: | 2025–07–22 |
| URL: | https://d.repec.org/n?u=RePEc:ehl:lserod:128950 |
| By: | Lin Li |
| Abstract: | Using an intangible intensity factor that is orthogonal to the Fama--French factors, we compare the role of intangible investment in predicting stock returns over the periods 1963--1992 and 1993--2022. For 1963--1992, intangible investment is weak in predicting stock returns, but for 1993--2022, the predictive power of intangible investment becomes very strong. Intangible investment has a significant impact not only on the MTB ratio (Fama--French high minus low [HML] factor) but also on operating profitability (OP) (Fama--French robust minus weak [RMW] factor) when forecasting stock returns from 1993 to 2022. For intangible asset-intensive firms, intangible investment is the main predictor of stock returns, rather than MTB ratio and profitability. Our evidence suggests that intangible investment has become an important factor in explaining stock returns over time, independent of other factors such as profitability and MTB ratio. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.16336 |
| By: | Eleonora Granziera; Vegard H. Larsen; Greta Meggiorini; Leonardo Melosi |
| Abstract: | We examine how speeches by Federal Open Market Committee (FOMC) members, including regional Fed presidents, shape private sector expectations. Speeches that signal rising inflationary pressures prompt both households and professional forecasters to raise their inflation expectations, consistent with Delphic effects. Only professional forecasters respond to Odyssean communications—statements about the Fed’s intended policy response—leaving Delphic effects as the dominant channel for households. These household responses are driven by speeches from regional presidents, likely due to greater visibility in regional media coverage. A general equilibrium model, featuring agents who differ in their ability to interpret Odyssean signals, explains this heterogeneity. |
| Keywords: | central bank communication, Delphic, Odyssean, inflation expectations, textual analysis, expectation formation |
| JEL: | E31 E58 D83 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_11992 |
| By: | Peter Haan (FU Berlin, DIW Berlin, Berlin School of Economics); Chen Sun (HU Berlin); Georg Weizsäcker (HU Berlin); Felix Weinhardt (European University Viadrina) |
| Abstract: | Different methods of eliciting long-run expectations yield data that predict economic choices differently well. We ask members of a wide population sample to make a 10-year investment decision and to forecast stock market returns in one of two formats: they either predict the average of annual growth rates over the next 10 years, or they predict the total, cumulative growth that occurs over the 10-year period. Results show that total 10-year forecasts are more pessimistic than average annual forecasts, but they better predict experimental portfolio choices and real-world stock market participation. |
| Keywords: | household finance; long-run predictions; survey experiments; |
| JEL: | D01 D14 D84 D9 |
| Date: | 2025–07–22 |
| URL: | https://d.repec.org/n?u=RePEc:rco:dpaper:539 |
| By: | Peter Haan; Chen Sun; Felix Weinhardt; Georg Weizsäcker |
| Abstract: | Different methods of eliciting long-run expectations yield data that predict economic choices differently well. We ask members of a wide population sample to make a 10-year investment decision and to forecast stock market returns in one of two formats: they either predict the average of annual growth rates over the next 10 years, or they predict the total, cumulative growth that occurs over the 10-year period. Results show that total 10- year forecasts are more pessimistic than average annual forecasts, but they better predict experimental portfolio choices and real-world stock market participation. |
| Keywords: | Household finance, long-run predictions, survey experiments |
| JEL: | D01 D14 D84 D9 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:diw:diwwpp:dp2130 |
| By: | Roberto Casarin (Ca' Foscari University of Venice); Antonio Peruzzi (Ca' Foscari University of Venice); Davide Raggi (Ca' Foscari University of Venice) |
| Abstract: | We study a New Keynesian Phillips curve in which agents deviate from the rational expectation paradigm and forecast inflation using a simple, potentially misspecified autoregressive rule. Consistency criteria à la Hommes and Zhu (2014) between perceived and actual laws of motion of inflation might allow for multiple expectational equilibria. Unfortunately, multiple equilibria models pose challenges for empirical validation. This paper proposes a latent Markov chain process to dynamically separate such equilibria. Moreover, an original Bayesian inference approach based on hierarchical priors is introduced, which naturally offers the possibility of incorporating equilibrium-identifying constraints with various degrees of prior beliefs. Finally, an inference procedure is proposed to assess a posteriori the probability that the theoretical constraints are satisfied and to estimate the equilibrium changes over time. We show that common prior assumptions regarding structural parameters favor the separation of equilibria, thereby making the Bayesian inference a natural framework for Markov–switching Phillips curve models. Empirical evidence obtained from observed inflation, output gap, and the consensus expectations from the Survey of Professional Forecasters supports multiple equilibria, and we find evidence of temporal variation in over- and under-reaction patterns, which, to the best of our knowledge, have not been previously documented. Specifically, we observe that agents tend to underreact to shocks when inflation is high and persistent, whereas they behave substantially as fully informed forecasters when the inflation level is low and stable, i.e., after the mid–nineties. We also find that the model does not suffer from the missing disinflation puzzle during the Great Recession. |
| Keywords: | Bounded rationality; Markov Switching; Multiple equilibria; Under-reaction; Bayesian methods; Horseshoe hierarchical priors; Survey of Professional Forecasters |
| JEL: | C11 C24 E31 D84 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ven:wpaper:2025:10 |