Högskolan i Skövde

With shorter product life cycles and increased competition, forecasting new product sales prior to launch is vital. We contribute to the new product forecasting literature by augmenting analogy-based approaches with pre-release online search traffic. In contrast to existing research, which relies solely on pre-release buzz during the launch phase, we consider life cycle sales. We propose a model of pre-release online search traffic and market potential, connecting the two to support pre-launch decision-making. We validate this relationship with an empirical experiment on sequential video game sales. Our findings support that pre-release online search traffic contains predictive information up to 18 weeks before release and can increase life cycle sales forecast accuracy by up to 21 %. The explanatory power of pre-release online search traffic varies across product generations. This evolution opens up marketing opportunities and highlights the importance of managing pre-release search interest. Our approach can be implemented with minimal data requirements, making it a versatile and accessible tool for firms. We provide extensive managerial findings and a way forward for incorporating this approach into “new to the world” products.

We complement the previous discussions of Bernanke’s review of the Bank of England’s forecasting activities and highlight directions for future research that are relevant to central banks and the wider forecasting community. Decisions in central banks, such as monetary policy ones, are hardly algorithmic and are often influenced by policy and current soft contextual information, introducing challenges into evaluating and specifying forecasts. The use of alternatives to standard econometric models is highlighted in the Bernanke report and other commentaries in this series. These methodological alternatives require both more research, to be validly applied and evaluated, and a cultural shift for those with forecasting responsibilities in central banks. Critically, uncertainty estimates in central bank forecasts are hardly purely model-based. How this is done and how to best communicate it to stakeholders and counterparties are fertile areas for research with potentially important implications for market participants. Finally, while academic research often focuses on large, well-funded central banks, there is a significant opportunity to help smaller, less-resourced institutions.

The provision of patient care in hospital pharmacies requires forecasting across a range of operations planning horizons to support adequate drug availability. This means that demand forecasts need to (1) be aligned across various planning horizons to support inventory management, (2) accommodate volatile drug delivery lead times and (3) be robust against erratic demand patterns including varying levels of intermittency. These are the challenges observed at the UK-based hospital pharmacy in the study. In response, we propose constructing forecasts that leverage the different time scales intrinsic to hospital pharmacies’ inventory management. Using temporal hierarchies, we mitigate the challenge of intermittency and volatility in drug demand, while also enabling coherent decisions across planning time scales. Across the range of service requirements, lead times, and from ‘Drug’ to ‘Drug by Dispensary’ planning, the proposed approach, based on temporal hierarchies, ranks consistently well, reducing modelling risk and supporting automation. It ranks statistically best on pinball loss when planning for 95–99% service at the Drug by Dispensary level for all lead times from 1 to 20 days, a key challenge for the hospital. The approach is model-agnostic allowing hospital pharmacies to adopt either state-of-the-art forecasting models, or adjust to existing software and modelling capabilities.

Leading indicators have been shown to be useful in predicting demand. Nowadays, a large number of potentially interesting variables are available in open databases. This makes it challenging to select the predictively relevant indicators. Lasso regression has become a popular choice when handling many variables, however, its selection is sensitive to changes in the sample and in the presence of correlated variables. This can harm both its predictive performance and the trustworthiness of the forecasts. We consider various approaches to aid variable selection. Specifically, we consider clustering via various statistical based methods, semantic information using a Semantic Bidirectional Encoder Representations from Transformers, or meta-data, such as a popularity index of variables. The resulting groups of variables are evaluated for selection directly, using sequential lasso, or transformed into cluster profiles, or factors using principal component analysis. Using an empirical case, we evaluate the alternative options on their decision and predictive performance, and on the interpretability of the resulting models. Our analysis indicates that there are trade-offs between predictive and decision performance, and interpretability. Semantic information is found to benefit variable selection, although we do not identify a dominant approach. Finally, we emphasise the need for methodological advances for explainability in forecasting. Highlights We compare different groupings of variables to enhance selection with lasso. Groupings are based on statistical, semantic, and meta-data information. Additional dimensionality reduction by cluster profiles and factors is investigated. Semantic information leads to gains in forecast and decision metrics. Credibly explainable models should match or outperform non-explainable benchmarks.

With historic data losing most of its value following shocks, this paper proposes a novel approach to combine counterfactual forecasts with a diffusion process to generate monthly recovery forecast of Chinese outbound tourism post-COVID-19. The counterfactual forecast uses a combination forecasting method based on temporal aggregation. Recovery rate forecasts follow a gradual diffusion process, used to scale the counterfactual forecast, resulting in the final prediction. The strength of this approach lies in its capacity to combine outputs from a conventional forecasting model with a theoretical framework that describes the evolution of a tourism destination as a sigmoid curve. It delivers remarkable accurate interval forecasts and offers a flexible framework that can be adapted to different contexts and decision horizons.

This paper introduces a methodology for incorporating risk preferences directly into forecasting model selection. The relative model information score, estimated from either a point-based information criterion or cross-validated errors, leverages the full distribution to map different risk propensities. We show that standard model selection in the literature is risk-agnostic. A risk-neutral stance is represented by the median of the relative model information score distribution, which characterises the plausibility of a model choice, while risk-averse and risk-tolerant choices correspond to its upper and lower quantiles. Our empirical evaluation demonstrates that risk-neutral and risk-averse selections consistently outperform the benchmark risk-agnostic choice in both point and quantile forecast accuracy. Moreover, we show that a risk-tolerant selection is beneficial during periods of extreme disruption. The proposed methodology provides a robust and flexible way to manage the forecast modelling risk, improving forecast accuracy and aligning forecasting modelling with stakeholders’ risk profiles.

PRACTITIONER SUMMARY: In this research we introduce a methodology for incorporating stakeholders’ risk preferences directly in the forecasting process. Given a degree of risk-aversion, or risk-tolerance, our methodology can guide the selection of the appropriate forecast that exhibits these risk characteristics in its errors. Risk-averse forecasts will minimise the probability of large errors, while risk-neutral forecasts will focus on minimising errors irrespective of their magnitude. The standard model selection statistics in the literature are typically risk-agnostic and do not enable the analyst to implement a preference. Using data from retailing, we provide evidence that both risk-neutral and risk-averse forecasts outperform the standard solutions both in terms of point and quantile accuracy (from 4% to 82% depending on the case), with longer forecast horizons benefiting from increased risk-aversion. Additionally, we demonstrate that risk-tolerant forecasts are useful when there are large disruptions, resulting in reduced forecast errors. Our methodology is robust with respect to the choice of the exact level of risk-preference, therefore reducing the need to precisely elucidate the risk profiles of stakeholders, which can be challenging in practice. The proposed approach can be operationalised with any statistical, machine learning, or judgmental forecasts, allowing direct incorporation of user preferences in model selection. Moreover, it does not require any changes in the forecasting models, simplifying its adoption in practice. Our work constitutes a first step in incorporating stakeholder risk preferences into the forecasting process, and therefore into the decisions supported by these forecasts.

We introduce the concept of temporal hierarchy forecasting (THieF) in predicting day-ahead electricity prices and show that reconciling forecasts for hourly products and 2- to 24-hour blocks can significantly (up to 13%) improve accuracy at all levels. These results remain consistent throughout a challenging 4-year test period (2021–2024) in the German and Spanish power markets and across model architectures, including linear regression, shallow feedforward neural networks, gradient-boosted decision trees, and a state-of-the-art, pretrained transformer. Given that ( i ) trading of block products is becoming more common and ( ii ) the computational cost of reconciliation is comparable to that of predicting hourly prices alone, we recommend using it in daily forecasting practice.

In operations of road freight transport, demand forecasts are mostly used for downstream optimization, following a predict then optimize setting. While the literature provides numerous optimization models that rely on accurate forecasts to support the planning of road freight transport, dedicated work on forecasting is limited, hindering a holistic treatment of the problem. Moreover, there is a disconnect to advances in the broader forecasting literature, limiting the adoption of modeling innovations and methodological advances. These can harm the quality and validity of forecasts designed to support road freight transport. We link the relevant forecasting publications to different prominent optimization problems for road freight transportation, highlighting disconnects between forecasting and optimization models in the area. By contrasting these with the current discourse in the forecasting literature, we identify relevant modeling and methodological improvements, in model building, the choice of loss function, and evaluation. These are important to better link forecasts with optimization models for road freight transport. Furthermore, we propose a unified hierarchical framework for freight demand to support aligned decisions across planning levels, which is an important consideration for practice. Our review helps structure and steer academic research and provides opportunities to improve existing forecasting models; a closer integration with optimization; and ultimately improve decisions.

Inventory management relies on accurate demand forecasts. Typically, these are univariate forecasts extrapolating patterns from past demand. The disaggregate nature of demand at the Stock Keeping Unit (SKU) level makes the incorporation of external information challenging. Nonetheless, such leading information can be critical to identifying disruptions and changes in the demand dynamics. To address the inventory planning needs of a global manufacturer we propose a methodology that identifies predictively useful leading indicators at an aggregate demand level, and translates that information to SKU-demand by leveraging on the hierarchical structure of the problem. Therefore, the proposed methodology provides probabilistic forecasts enriched by leading indicator information at SKU-level, as inputs for inventory management. The methodology automatically adjusts the choice of indicators for different required lead times, with some being more informative about the short-term demand dynamics and others for the long-term. We demonstrate the benefits both in the case of backorders and lost-sales, for a variety of lead times. We further benchmark the solution against solely using leading indicators or hierarchical forecasts, demonstrating that the benefits appear primarily by the proposed blending of the modelling approaches. The outcome is demonstratively better forecasts and inventory management for the case company. Additionally, management gains insights into the main drivers of their short and long-term demand, and the ability to adjust inventory replenishment accordingly. The ability to account for diverse macro and market information in operations is paramount for firms with a global reach that face different market conditions across countries. Additionally, the transparency of which leading indicators are influencing forecasts of different lead times is conducive to increased forecast trustworthiness. 

The Bullwhip Effect, the magnification of demand variability throughout the supply chain, poses a challenge to firms. Inaccurate forecasts increase it, with forecast errors translating into higher inventory costs at a local level and impacting other members of the supply chain, as their decisions are based on mis-estimated incoming orders. The conventional measure for the Bullwhip Effect does not reflect how forecast uncertainty evolves in the supply chain. A new metric is proposed that overcomes many of the limitations of the Bullwhip Ratio: the Ratio of Forecast Uncertainty. It benchmarks the upstream forecast errors to the downstream's. An inventory simulation is deployed to study the properties and usefulness of this measure. It connects to inventory costs at the upstream level and holds more explanatory power than the standard Bullwhip Ratio and the complementary Net Stock Amplification. Managers can use it to better understand the upstream impact of the forecasting process.