Multistep quantile forecasts for supply chain and logistics operations: bootstrapping, the GARCH model and quantile regression based approaches

Abstract

In this paper, we discuss and compare empirically various ways of computing multistep quantile forecasts of demand, with a special emphasis on the use of the quantile regression methodology. Such forecasts constitute a basis for production planning and inventory management in logistics systems optimized according to the cycle service level approach. Different econometric methods and models are considered: direct and iterated computations, linear and nonlinear (GARCH) models, simulation and nonsimulation based procedures and parametric as well as semiparametric specifications. These methods are applied to compute multiperiod quantile forecasts of the monthly microeconomic time series from the popular M3 competition database. According to various accuracy measures for quantile predictions, the best procedures are based on simulation techniques using predictive distributions generated by either the quantile regression methodology combined with random draws from the uniform distribution or parametric and nonparametric bootstrap techniques. These methods lead to large reductions in the total costs of logistics systems as compared with non-simulation based procedures. For example, in the case of forecasting 12 months ahead, relatively short time series and a high cycle service level, the quantile regression based simulation approach reduces the average supply chain cost per unit of output by about 70-85%. At the shortest horizons, the GARCH model should be seriously considered among the preferred forecasting solutions for production and inventory planning.

Optimization Design of Short Life Cycle Product Logistics Supply Chain Scheme Based on Support Vector Machine

“…is article studies the win-win problem of supply chain channels achieved by manufacturers through quantity discount contracts in a mixed channel supply chain systems [9,10].…”

Research on Freight Development of Guangdong Province Based on Grey Theory Model

“…In recent years, logistics development prediction has become a hot issue for scholars. e methods commonly used by scholars include exponential smoothing [5][6][7], linear model [8], BP neural network method [5][6][7][8][9], multiple regression analysis [5,9], seasonal autoregressive model [10][11][12][13], discrete wavelet technology [14], vector autoregressive method [15], and Markov chain theory [16], which are widely used in the prediction of logistics and freight development. In addition to conventional prediction methods, scholars also innovate prediction methods, such as a genetic algorithm and backpropagation (GA-BP) prediction model (optimized backpropagation neural network model using genetic algorithm), which are used to predict freight volume demand with small error [17].…”

ParAMS: Parameter Optimization for Atomistic and Molecular Simulations

“…In our experience, the majority of researchers, although being interested in individual parameter fitting, are discouraged by the high barrier that comes with it. The main reason for this being a lack of generalization and transparency: (1) Training data often come in a variety of formats. (2) Optimizers expect a different input all together.…”

“…Throughout the years, the use of predictive computational models has become standard practice in many professional fields such as logistics, economics, and R&D, − with computational chemistry being no exception. Predictive models in this field often approximate the potential energy surface (PES) and derived properties for a given chemical system and can be broadly categorized based on their level of theory; quantum mechanical (QM) approaches such as wave function or density functional theory explicitly model the electronic structure, generally resulting in a high accuracy and broad applicability.…”