Research

Abstract: 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.

Abstract: 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.

Abstract: Forecasting student performance can help to identify students at risk and aids in recommending actions to improve their learning outcomes. That often involves elaborate machine learning pipelines. These tend to use large feature sets including behavioral data from learning management systems or demographic information. However, this complexity can lead to inaccurate predictions when concept drift occurs, or when a large number of features are used with a limited sample size.

We investigate the performance of different machine learning pipelines on a data set with change in study behavior during the Covid-19 period. We demonstrate that (i) LASSO, a shrinkage estimator that reduces complexity and overfitting, outperforms several machine learning models under these circumstances, (ii) a linear regression relying on only two handcrafted features achieves higher accuracy and substantially less predictive bias than commonly used, more complex models with large feature sets. Due to their simplicity, these models can serve as a benchmark for future studies and a fallback model when substantial concept or covariate drift is encountered.