Point Forecasts Research Articles

Mining Google Trends data for nowcasting and forecasting colorectal cancer (CRC) prevalence.

Colorectal cancer (CRC) is the third most prevalent and second most lethal form of cancer in the world. Consequently, CRC cancer prevalence projections are essential for assessing the future burden of the disease, planning resource allocation, and developing service delivery strategies, as well as for grasping the shifting environment of cancer risk factors. However, unlike cancer incidence and mortality rates, national and international agencies do not routinely issue projections for cancer prevalence. Moreover, the limited or even nonexistent cancer statistics for large portions of the world, along with the high heterogeneity among world nations, further complicate the task of producing timely and accurate CRC prevalence projections. In this situation, population interest, as shown by Internet searches, can be very important for improving cancer statistics and, in the long run, for helping cancer research. This study aims to model, nowcast and forecast the CRC prevalence at the global level using a three-step framework that incorporates three well-established univariate statistical and machine-learning models. First, data mining is performed to evaluate the relevancy of Google Trends (GT) data as a surrogate for the number of CRC survivors. The results demonstrate that population web-search interest in the term "colonoscopy" is the most reliable indicator to nowcast CRC disease prevalence. Then, various statistical and machine-learning models, including ARIMA, ETS, and FNNAR, are trained and tested using relevant GT time series. Finally, the updated monthly query series spanning 2004-2022 and the best forecasting model in terms of out-of-sample forecasting ability (i.e., the neural network autoregression) are utilized to generate point forecasts up to 2025. Results show that the number of people with colorectal cancer will continue to rise over the next 24 months. This in turn emphasizes the urgency for public policies aimed at reducing the population's exposure to the principal modifiable risk factors, such as lifestyle and nutrition. In addition, given the major drop in population interest in CRC during the first wave of the COVID-19 pandemic, the findings suggest that public health authorities should implement measures to increase cancer screening rates during pandemics. This in turn would deliver positive externalities, including the mitigation of the global burden and the enhancement of the quality of official statistics.

A multiplicative Holt–Winters model and autoregressive moving-average for hyponatremia mortality rates

This study presents the effective utilization of data analysis in efficiently managing medical resources — a crucial factor in ensuring high-quality healthcare delivery and positive patient outcomes. We employ a multiplicative Holt-Winters model and a linear model with an autoregressive moving-average time series error term and analyze time series mortality rates associated with hyponatremia, a condition that leads to a systemic electrolyte imbalance. The time series data consistently reveals noticeable trends and recurring seasonal patterns. To address these intricate patterns, the proposed data analysis approaches adeptly incorporate these features, thereby yielding relatively accurate point forecasts and reliable forecast intervals. Beyond mere forecasting, the autoregressive moving-average model approach offers insights into the overarching trend and facilitates the comparison of seasonal components. One of the approaches is well-suited for the specific time series related to hyponatremia and holds the potential for extension to incorporate more complex trends beyond linearity. The research demonstrates the synergistic relationship between data analysis and medical expertise, working in tandem to achieve the optimal allocation of medical resources. We strongly advocate for the routine integration of statistical analyses as an essential practice to foster this harmonious synergy.

POST-BaLSTM: A Bagged LSTM forecasting ensemble embedded with a postponement framework to target the semiconductor shortage in the automotive industry

In the automotive business environment, with intense competition among firms and with the added element of what is known as the “chips or semiconductor shortage,” adequate demand and supply chain planning is not forthright as semiconductors are essential to fabricating printed circuit boards, which in turn are necessary to control numerous computerized requirements of the vehicle. This paper presents a manufacturing postponement framework concerted with an ensemble forecasting method based on multi-layer Long Short-Term Memory networks combined with a bagging approach to predict highly fluctuating non-stationary printed circuit boards demand data in a business offering electronics manufacturing services to the automotive industry. The devised postponement framework effectively addresses the impending modifications required by the company to cope with the shortage of semiconductors. When the proposed forecasting method is embedded with the postponement framework, the procedure is called the Postponement Bagged Long Short-Term Memory (POST-BaLSTM) model. The approach aims to optimize semiconductor use, minimize wastage, and facilitate effective resource planning. The proposed method automatically fits the forecasts into the postponement structure by considering several Long Short-Term Memory models previously optimized using hyperparameters grid search. Bagged point forecasts and prediction intervals are derived using model estimates of quantile calculations. In our implementation, we demonstrate that a two-layered archetype using our method can lead to nearly a 43% improvement in measuring forecast errors compared to statistical and machine learning methods. Moreover, our approach bestows a 75% rating interval accuracy when measured with the denoted Spectrum Fidelity assessment.

A novel selective ensemble system for wind speed forecasting: From a new perspective of multiple predictors for subseries

Wind speed forecasting is of considerable economic and social significance; however, it remains challenging. Most state-of-the-art methods attempt to select the optimal predictor for each subseries in the framework of decomposition-ensemble forecasting; however, these do not always produce desirable forecasting results. Hence, this study innovatively proposes the idea of the validity of the forecasted subseries and considers the significance of the diversity of subseries predictors. In this context, a novel selective ensemble wind speed forecasting system is proposed from the new perspective of multiple predictors for subseries, which is advantageous owing to its data pretreatment, multiple predictors, adaptive selection, and multi-objective ensemble modules. Particularly, the data pretreatment module is designed based on successive variational mode decomposition, which can settle the limitations of the original algorithm as well as obtain desirable subseries. Meanwhile, the multiple predictors module is developed to provide a variety of predictors for each subseries by introducing the extreme learning machine series models. The adaptive selection module is designed to identify the effective subseries forecasted by multiple predictors module, whereas the multi-objective ensemble module is proposed to realize the wind speed point and interval forecasting based on a multi-objective optimization algorithm, which resolves the limitation of a single subseries predictor and further improves the power of the developed wind speed forecasting system. The interesting finding is that for the same dataset, the valid predictors of different subseries are different; furthermore, for a subseries, there may be multiple or no valid predictors. Experiments based on three datasets from the Chengde wind farm indicate that the mean absolute percentage errors of the wind speed point forecasting of the proposed system are 1.763238%, 2.169993%, and 1.578143%, respectively. For interval forecasting, when the confidence level is 90%, prediction interval coverage probability are 100.000000, 100.000000, and 99.758454, respectively. Hence, the proposed selective ensemble system can be employed as a novel solution for wind speed forecasting.

Causal carbon price interval prediction using lower upper bound estimation combined with asymmetric multi-objective evolutionary algorithm and long short-term memory

Recently the interval forecasting of carbon price is investigated by advanced research since it can better quantify the uncertainty and reliability of the forecast value in comparison with point forecasting. However, this kind of model is always limited to the distribution-based method, which can only conduct symmetric prediction intervals and heavily rely on accurate point prediction and preprogrammed errors’ probability distribution. Therefore, we make the following improvements for enriching research on the interval prediction of carbon price and providing more references for researchers in related fields. Firstly, lower upper bound estimation model (LUBE) is employed to conduct the prediction interval rather than the distribution-based method, and this designer can generate asymmetric upper and lower bounds without the need for assumptions and humanly devised parameters. Secondly, the causal inference is applied to the feature selection rather than the correlation analysis, which can well improve the generalization ability of prediction model. Thirdly, we improved the objectives when using multi-objective evolutionary algorithms for searching more efficient solutions in less time. Finally, through various experiments the most effective combination of LUBE and the ensemble method are verified, which are also proved by experiments.

Modeling the out-of-sample predictive relationship between equity premium, returns on the price of crude oil and economic policy uncertainty using multivariate time-varying dimension models

Researchers increasingly rely on the newspaper-based uncertainty (volatility) measures pioneered by Baker et al. (2016) to forecast economic, financial variables and commodity prices out-of-sample. Among them, equity premium and returns on the price of crude oil have received a great deal of attention given their importance. By combining different linear state-space representations of the multivariate dynamic linear model with the discount factor-based model averaging (selection) technique outlined in Raftery et al. (2010), and using monthly data from 1997m1 through 2022m10, we suggest three multivariate time-varying dimension models, and forecast these variables out-of-sample in a contemporaneous fashion. The time-varying dimension feature allows the number of predictors in each regression of the multivariate system to change over time. From a technical viewpoint, the suggested models are intuitive (flexible), and do not require much subjective input from the researcher. They also produce very accurate one-month ahead out-of-sample density (point) forecasts on average. From an empirical viewpoint, our analysis provides new and interesting insights into the out-of-sample predictive relationship between equity premium, returns on the price of crude oil and economic policy uncertainty.

A combined prediction system for PM2.5 concentration integrating spatio-temporal correlation extracting, multi-objective optimization weighting and non-parametric estimation

Air pollution nowadays has seriously hindered the sustainable development. PM2.5 greatly affects air quality and human health, even facilitates virus transmission, making its concentration prediction is crucial. However, previous studies are limited to single PM2.5 concentration series, neglecting optimization of prediction stability, and lacking uncertainty analysis. To address these issues, this research proposed a combined PM2.5 prediction system (CPPs) based on modular concept. Firstly, the temporal and spatial correlations of PM2.5 were fully extracted by data pretreatment and feature selection modules. Subsequently, the results of single submodel prediction module were integrated by multi-objective slime mould algorithm in combination weighted module, achieving Pareto optimality theoretically. Eventually, interval forecasting module analyzed the predictive uncertainty. Notably, a truly accurate metric for predictive directional accuracy was proposed for the first time. Validating CPPs using PM2.5 data from Shanghai during COVID-19 epidemic showed superior performance in point and interval forecasts. This research achieves optimization of prediction accuracy and stability as well as uncertainty analysis based on multiple data sources, contributes to improved air quality and public health protection.

Wasserstein barycenter for link prediction in temporal networks

Abstract We propose a flexible link forecast methodology for weighted temporal networks. Our probabilistic model estimates the evolving link dynamics among a set of nodes through Wasserstein barycentric coordinates arising within the optimal transport theory. Optimal transport theory is employed to interpolate among network evolution sequences and to compute the probability distribution of forthcoming links. Besides generating point link forecasts for weighted networks, the methodology provides the probability that a link attains weights in a certain interval, namely a quantile of the weights distribution. We test our approach to forecast the link dynamics of the worldwide Foreign Direct Investments network and of the World Trade Network, comparing the performance of the proposed methodology against several alternative models. The performance is evaluated by applying non-parametric diagnostics derived from binary classifications and error measures for regression models. We find that the optimal transport framework outperforms all the competing models when considering quantile forecast. On the other hand, for point forecast, our methodology produces accurate results that are comparable with the best performing alternative model. Results also highlight the role played by model constraints in the determination of future links emphasising that weights are better predicted when accounting for geographical rather than economic distance.

Improved Analog Ensemble Formulation for 3-Hourly Precipitation Forecasts

Abstract Analog ensembles (AnEns) traditionally use a single numerical weather prediction (NWP) model to make a forecast, then search an archive to find a number of past similar forecasts (analogs) from that same model, and finally retrieve the actual observations corresponding to those past forecasts to serve as members of an ensemble forecast. This study investigates new statistical methods to combine analogs into ensemble forecasts and validates them for 3-hourly precipitation over the complex terrain of British Columbia, Canada. Applying the past analog error to the target forecast (instead of using the observations directly) reduces the AnEn dry bias and makes prediction of heavy-precipitation events probabilistically more reliable—typically the most impactful forecasts for society. Two variants of this new technique enable AnEn members to obtain values outside the distribution of the finite archived observational dataset—that is, they are theoretically capable of forecasting record events, whereas traditional analog methods cannot. While both variants similarly improve heavier precipitation events, one variant predicts measurable precipitation more often, which enhances accuracy during winter. A multimodel AnEn further improves predictive skill, albeit at higher computational cost. AnEn performance shows larger sensitivity to the grid spacing of the NWP than to the physics configuration. The final AnEn prediction system improves the skill and reliability of point forecasts across all precipitation intensities. Significance Statement The analog ensemble (AnEn) technique is a data-driven method that can improve local weather forecasts. It improves raw model forecasts using past similar model predictions and observations, reducing future forecast errors and providing probabilities for a range of possible outcomes. One limitation of AnEns is that they commonly tend to make rare-event (e.g., heavy precipitation) forecasts appear less extreme. Usually, heavier precipitation events have a higher impact on society and the economy. This study introduces two new AnEn techniques that make operational forecasts of both probabilities and most likely amounts more accurate for heavy precipitation.

Improved Sieve Bootstrap based prediction intervals for time series

Forecasting future observations is one of the main goals of time series modeling. Point forecasts are the most common type of prediction, but interval predictions are more informative, usually obtained as prediction intervals (PIs). The conventional Box-Jenkins PIs perform well when the underlying distributional assumptions hold. To circumvent such adherence, resampling methods such as residual based Bootstrap were one of the approaches for constructing PIs, but these are only suitable when the order of the process is known. Sieve Bootstrap method has the advantage of requiring no such prior knowledge. In this work, two new Sieve Bootstrap methods have been proposed. The performance of the proposed methods for constructing PIs have been compared with the existing Bootstrap methods. One of the proposed methods, namely, predictive residual based rescaled Sieve Bootstrap (PRSB) has been found to outperform the other proposed predictive residual based Sieve Bootstrap (PSB) and also the existing methods considered. PSB method has been found to be superior to Box-Jenkins and existing Sieve Bootstrap methods. The proposed methods are not only an improvement over the existing predictive residual based Bootstrap methods in AR(p) set up but also have been extended to be applicable in ARMA(p, q) situations as well.

An approach for prospective forecasting of rock slope failure time

Rock slope failures globally account for most single-event landslide disasters. Climatic changes in mountain areas boost failure activity and the demand for reliable failure time forecasts. State-of-the-art prediction models are often confused with high-frequency slope deformation data. Prospectively, they provide ambiguous forecasts as data filtering, starting point definition and forecast uncertainty remain arbitrary. Here, we develop a prospective failure time forecast model that applies multiple filtering and inverse velocity percentiles to minimize subjective decisions. We test the concept with 14 historic slope failures of 102-108 m3 including 46 displacement datasets from different sensors. After automatic detection of the onset of acceleration, the failure time of all events is forecasted to within −1 ± 17 h for higher-frequency data and −1 ± 4 d for daily data with a final mean uncertainty of 1 ± 1 d and 7 ± 4 d that is estimated in real-time. This prospective approach overcomes previous long-standing problems by introducing a robust and uniform concept across various types of catastrophic slope failures and sensors.

When will Arctic sea ice disappear? Projections of area, extent, thickness, and volume

Rapidly diminishing Arctic summer sea ice is a strong signal of the pace of global climate change. We provide point, interval, and density forecasts for four measures of Arctic sea ice: area, extent, thickness, and volume. Importantly, we enforce the joint constraint that these measures must simultaneously arrive at an ice-free Arctic. We apply this constrained joint forecast procedure to models relating sea ice to atmospheric carbon dioxide concentration and models relating sea ice directly to time. The resulting “carbon-trend” and “time-trend” projections are mutually consistent and predict a nearly ice-free summer Arctic Ocean by the mid-2030s with an 80% probability. Moreover, the carbon-trend projections show that global adoption of a lower carbon path would likely delay the arrival of a seasonally ice-free Arctic by only a few years.

Temporal Data Pooling With Meta-Initialization for Individual Short-Term Load Forecasting

The growth of advanced metering infrastructure (AMI) deployment has enabled intelligent power control and management at the customer level; Highly accurate individual short-term load forecasting is crucial for this precise control per customer. Recently, deep learning has been widely adopted to improve forecasting accuracy. However, training an individual deep network (local training) has overfitting issues due to data paucity per customer. Thus, a pooling scheme has been introduced to augment a training dataset by batching several customers’ data. Nevertheless, there is room for existing pooling approaches to further improve accuracy by considering distribution heterogeneity within a customer dataset. In addition, their static pool assignment only with a customer’s training dataset may cause accuracy degradation under concept drift in serving time. To overcome these, we propose a Temporal Data Pooling (TDP) that constructs data pools at the data sample level with a novel distribution inference method and theoretical analysis. It allows the most probable forecasting model to serve predictions while resolving data shortage issues in local training. The TDP outperforms the other six competing methods for point and probabilistic forecasting; it shows robust accuracy under concept drift. Moreover, it demonstrates superior accuracy for unseen customers without additional training, proving its scalability.

Forecasting intraday financial time series with sieve bootstrapping and dynamic updating

AbstractIntraday financial data often take the form of a collection of curves that can be observed sequentially over time, such as intraday stock price curves. These curves can be viewed as a time series of functions observed on equally spaced and dense grids. Due to the curse of dimensionality, high‐dimensional data pose challenges from a statistical aspect; however, it also provides opportunities to analyze a rich source of information so that the dynamic changes within short‐time intervals can be better understood. We consider a sieve bootstrap method to construct 1‐day‐ahead point and interval forecasts in a model‐free way. As we sequentially observe new data, we also implement two dynamic updating methods to update point and interval forecasts for achieving improved accuracy. The forecasting methods are validated through an empirical study of 5‐min cumulative intraday returns of the S&P/ASX All Ordinaries Index.

Time to build a new practice of foresight for national economies? Ireland, and uncertain futures in forecasts and scenarios

PurposeThe literature on economic forecasting, is showing an increase in criticism, of the inaccuracy of forecasts, with major implications for economic, and fiscal policymaking. Forecasts are subject to the systemic uncertainty of human systems, considerable event-driven uncertainty, and show biases towards optimistic growth paths. The purpose of this study is to consider approaches to improve economic foresight.Design/methodology/approachThis study describes the practice of economic foresight as evolving in two separate, non-overlapping branches, short-term economic forecasting, and long-term scenario analysis of development, the latter found in studies of climate change and sustainability. The unique case of Ireland is considered, a country that has experienced both steep growth and deep troughs, with uncertainty that has confounded forecasting. The challenges facing forecasts are discussed, with brief review of the drivers of growth, and of long-term economic scenarios in the global literature.FindingsEconomic forecasting seeks to manage uncertainty by improving the accuracy of quantitative point forecasts, and related models. Yet, systematic forecast failures remain, and the economy defies prediction, even in the near-term. In contrast, long-term scenario analysis eschews forecasts in favour of a set of plausible or possible alternative scenarios. Using alternative scenarios is a response to the irreducible uncertainty of complex systems, with sophisticated approaches employed to integrate qualitative and quantitative insights.Research limitations/implicationsTo support economic and fiscal policymaking, it is necessary support advancement in approaches to economic foresight, to improve handling of uncertainty and related risk.Practical implicationsWhile European Union Regulation (EC) 1466/97 mandates pursuit of improved accuracy, in short-term economic forecasts, there is now a case for implementing advanced foresight approaches, for improved analysis, and more robust decision-making.Social implicationsBuilding economic resilience and adaptability, as part of a sustainable future, requires both long-term strategic planning, and short-term policy. A 21st century policymaking process can be better supported by analysis of alternative scenarios.Originality/valueTo the best of the authors’ knowledge, the article is original in considering the application of scenario foresight approaches, in economic forecasting. The study has value in improving the baseline forecast methods, that are fundamental to contemporary economics, and in bringing the field of economics into the heart of foresight.

Density forecasting of conjunctivitis burden using high-dimensional environmental time series data

As one of the most common eye conditions being presented at clinics, acute conjunctivitis puts substantial strain on primary health resources. To reduce this public health burden, it is important to forecast and provide forward guidance to policymakers by estimating conjunctivitis trends, taking into account factors which influence transmission. Using a high-dimensional set of ambient air pollution and meteorological data, this study describes new approaches to point and probabilistic forecasting of conjunctivitis burden which can be readily translated to other infectious diseases. Over the period of 2012 – 2022, we show that simple models without environmental data provided better point forecasts but the more complex models which optimized predictive accuracy and combined multiple predictors demonstrated superior density forecast performance. These results were shown to be consistent over periods with and without structural breaks in transmission. Furthermore, ecological analysis using post-selection inference showed that increases in SO2, O3 surface concentration and total precipitation were associated to increased conjunctivitis attendance. The methods proposed can provide rich and informative forward guidance for outbreak preparedness and help guide healthcare resource planning in both stable periods of transmission and periods where structural breaks in data occur.

Projections and Epidemiology of Revision Hip and Knee Arthroplasty in the United States to 2040-2060

National projections of future joint arthroplasties are useful in understanding the changing burden of surgery and related outcomes on the health system. The aim of this study is to update the literature by producing Medicare projections for revision total joint arthroplasty procedures from 2040 through 2060.The study uses 2000-2019 data from the CMS Medicare Part-B National Summary and combines procedure counts using CPT codes for revision total joint arthroplasty procedures. In 2019, revision total knee arthroplasty (rTKA) and revision total hip arthroplasty (rTHA) procedures totaled 53,217 and 30,541, respectively, forming a baseline from which we generated point forecasts between 2020 and 2060 and 95% forecast intervals (FI).On average, the model projects an annual growth rate of 1.77% for rTHAs and 4.67% for rTKAs. By 2040, rTHAs were projected to be 43,514 (95% FI = 37,429-50,589) and rTKAs were projected to be 115,147 (95% FI = 105,640-125,510). By 2060, rTHAs was projected to be 61,764 (95% FI = 49,927-76,408) and rTKAs were projected to be 286,740 (95% FI = 253,882-323,852).Based on 2019 total volume counts, the log-linear exponential model forecasts an increase in rTHA procedures of 42% by 2040 and 101% by 2060. Similarly, the estimated increase for rTKA is projected to be 149% by 2040 and 520% by 2060. An accurate projection of future revision procedure demands is important to understand future healthcare utilization and surgeon demand. This finding is only applicable to the Medicare population and demands further analysis for other population groups.

A point and interval forecasting of solar irradiance using different decomposition based hybrid models

A point and interval forecasting of solar irradiance using different decomposition based hybrid models

Introducing shrinkage in heavy-tailed state space models to predict equity excess returns

AbstractWe forecast excess returns of the S &P 500 index using a flexible Bayesian econometric state space model with non-Gaussian features at several levels. More precisely, we control for overparameterization via global–local shrinkage priors on the state innovation variances as well as the time-invariant part of the state space model. The shrinkage priors are complemented by heavy tailed state innovations that cater for potential large breaks in the latent states, even if the degree of shrinkage introduced is high. Moreover, we allow for leptokurtic stochastic volatility in the observation equation. The empirical findings indicate that several variants of the proposed approach outperform typical competitors frequently used in the literature, both in terms of point and density forecasts.

A novel similar-day based probability density forecasting framework for residential loads

Probabilistic load forecasting can provide much richer information about uncertainties in residential loads than point load forecasting, and is gaining increasing attention in building efficient and economical demand response programs. However, obtaining satisfactory forecasting results for residential loads is challenging due to their intrinsic nonlinearity and volatility. Furthermore, existing probabilistic forecasting methods are limited in their ability to directly generate probability density forecasting in an efficient and accurate way. To address these issues, a new probability density forecasting framework for residential loads is proposed in this paper. Firstly, a similar-day choosing method is proposed by combining weighted external features and internal load curves to build a similar-day dataset. Then, point forecasting is achieved by training an exclusive predictor on the corresponding similar-day dataset for each forecasting target. Finally, probability density forecasting is realized by conducting kernel density estimation on the similar-day dataset and calibrating it according to the point forecasting result. Comparisons with other state-of-the-art methods carried out on the HUE dataset demonstrate the efficiency and superiority of the proposed framework in both point and probability density forecasting.