Search results

A method is developed for evaluating forecasting models with respect to both error and complexity in forecasting. Several types of forecasting accuracy measures (MSE, MPE, MAPE, Theil′s U‐Statistic and a loss cost function) are examined and the approach is illustrated using short‐term forecasting methods, and weekly and four‐weekly data. The approach can, however, be applied equally to immediate, medium‐ and long‐term forecasting.

Demand foresting significantly impacts supply chain (SC) design and recovery planning. The more accurate the demand forecast, the better the recovery plan and the more resilient the SC. Given the paucity of research about machine learning (ML) applications and the pharmaceutical industry’s need for disruptive techniques, this study aims to investigate the applicability and effect of ML algorithms on demand forecasting. More specifically, the study identifies machine learning algorithms applicable to demand forecasting and assess the forecasting accuracy of using ML in the pharmaceutical SC.

This research used a single-case explanatory methodology. The exploratory approach examined the study’s objective and the acquisition of information technology impact. In this research, three experimental designs were carried out to test training data partitioning, apply ML algorithms and test different ranges of exclusion factors. The Konstanz Information Miner platform was used in this research.

Based on the analysis, this study could show that the most accurate training data partition was 80%, with random forest and simple tree outperforming other algorithms regarding demand forecasting accuracy. The improvement in demand forecasting accuracy ranged from 10% to 41%.

This study provides practical and theoretical insights into the importance of applying disruptive techniques such as ML to improve the resilience of the pharmaceutical supply design in such a disruptive time.

The finding of this research contributes to the limited knowledge about ML applications in demand forecasting. This is manifested in the knowledge advancement about the different ML algorithms applicable in demand forecasting and their effectiveness. Besides, the study at hand offers guidance for future research in expanding and analyzing the applicability and effectiveness of ML algorithms in the different sectors of the SC.

Accounting and decision making rely heavily on forecasts. For several reasons, we should expect ongoing increases in forecasting accuracy. The purpose of this paper is to test the hypothesis of improved forecasts over time.

The paper analyzes original monthly sales plans and current data for three different car models in six different countries over 15 years and over several product life cycles (PLCs). Forecasting accuracy is calculated as one minus forecasting error. Forecasting error is measured with MAD/MEAN for periods of years or relative deviations per month. The hypothesis of decreasing forecasting errors is tested with the non‐parametric Mann/Kendall trend test. Additional interviews with managers were conducted to elicit details of internal forecasting organization and instruments.

The paper finds no evidence of increased forecasting accuracy in general over 15 years or over subsequent PLCs. This seems surprising, given improved statistical methods and software in general, and experience and learning effects of the organization itself. However, there is evidence from the case, that the reason lies in environmental uncertainty and volatility and not in internal factors within the control of the company.

Evidence from one case study is limited in its external validity. Future studies should analyze the forecasts of more companies, more industries and different forecasting objects, the latter including consumer, industrial goods and services. In the absence of further research, the results seem to negate the common assumption, that companies are generally able to make accurate forecasts, including those for accounting purposes. This hypothesis is clearly confuted.

The paper describes a methodology for companies to analyze their own forecasting accuracy and to identify possible reasons for a lack of accuracy, or basic approaches to increasing it.

Most studies on forecasting accuracy rely on interviews and questionnaires, entailing bias that is difficult to control. Few studies analyze archival data in order to measure forecasting accuracy; so that our study avoids much of the bias mentioned above. Despite the inevitable limitations of case studies, a study such as the present one at least allows us to dispute a common hypothesis about forecasting accuracy in practice.

The pace of booking is a critical element in the accuracy of revenue management (RM) systems. Anecdotal evidence suggests that booking windows exhibit persistent shifts due to a variety of macro and micro factors. The article outlines several causes and tests the impact of the shifts on forecasting accuracy.

A novel methodological approach is utilized to empirically shift hotel reservation windows into smaller increments. Forecasts are then estimated and tested on the incremental shifts with popular RM techniques characteristic of advance booking data. A random effects model assesses the impact of the shifts on forecast accuracy.

The results show that shifts in booking behavior can cause the accuracy of forecasting models to deteriorate. The findings stress the importance of considering these shifts in model estimation and evaluation.

The results demonstrate that changes in booking behavior can be detrimental to the accuracy of RM forecasting algorithms. It is recommended that revenue managers monitor booking window shifts when forecasting with advanced booking data.

This study is the first to systematically assess the impact of booking window shifts on forecasting accuracy. The demonstrated approach can be implemented in future research to assess model accuracy as booking behavior changes.

Financial analysts' forecasts serve as a proxy for market earnings expectations, and research provides mixed evidence of the relation between financial analysts' expertise and forecast accuracy. The judgment and decision-making (J/DM) literature suggests that those with more expertise will not perform better when tasks exhibit either extremely high or extremely low complexity. Expertise is expected to contribute to superior performance for tasks between these two extremes. Using archival data, this research examines the effect of analysts' expertise on forecasting performance by taking into consideration the forecasting task's complexity. Results indicate that expertise is not an explanatory factor for forecast accuracy when the forecasting task's complexity is extremely high or low. However, when task complexity falls between these two extremes, expertise is a significant explanatory variable of forecast accuracy. Both results are consistent with our expectations.

This paper aims to improve the accuracy of hotel daily occupancy forecasts – an essential element in the revenue management cycle – by proposing and testing a novel approach. The authors add the hotel competitive-set’s predicted occupancy as an input of the individual property forecast and, using a recursive approach, demonstrate that there is a potential for significant reduction in the forecasting error.

The paper outlines the theoretical justification and the mechanism for this new approach. It applies a simulation for exploring the potential to improve the accuracy of the hotel’s daily occupancy forecasts, as well as analysis of data from a field study of two hotel clusters’ daily forecasts to provide empirical support to the procedure’s viability.

The results provide strong support to the notion that the accuracy could be enhanced. Incorporating the competitive set prediction by using either a genetic algorithm or the simple linear regression model improves the accuracy of the forecast using either the absolute or the absolute percentage as the error measure.

The proliferation of data sharing practices in the hotel industry reveals that the timely data sharing-aggregation-dissemination mechanism required for implementing this forecasting paradigm is feasible.

Given the crucial role of accurate forecasts in revenue management and recent changes in the hotels’ operating environment which made it harder to achieve or maintain high levels of accuracy, this study’s proposed novel approach has the potential to make a unique contribution in the realm of forecasting daily occupancies.

In the last few decades, there has been growing interest in forecasting with computer intelligence, and both fuzzy time series (FTS) and artificial neural networks (ANNs) have gained particular popularity, among others. Rather than the conventional methods (e.g., econometrics), FTS and ANN are usually thought to be immune to fundamental concepts such as stationarity, theoretical causality, post-sample control, among others. On the other hand, a number of studies significantly indicated that these fundamental controls are required in terms of the theory of forecasting, and even application of such essential procedures substantially improves the forecasting accuracy. The aim of this paper is to fill the existing gap on modeling and forecasting in the FTS and ANN methods and figure out the fundamental concepts in a comprehensive work through merits and common failures in the literature. In addition to these merits, this paper may also be a guideline for eliminating unethical empirical settings in the forecasting studies.

Given that the use of Google Trends data is helpful to improve forecasting performance, this study aims to investigate whether the precision of forecast combination can benefit from the use of Google Trends Web search index along with the encompassing set.

Grey prediction models generate single-model forecasts, while Google Trends index serves as an explanatory variable for multivariate models. Then, three combination sets, including sets of univariate models (CUGM), all constituents (CAGM) and constituents that survive the forecast encompassing tests (CSET), are generated. Finally, commonly used combination methods combine the individual forecasts for each combination set.

The tourism volumes of four frequently searched-for cities in Taiwan are used to evaluate the accuracy of three combination sets. The encompassing tests show that multivariate grey models play a role to be reckoned with in forecast combinations. Furthermore, the empirical results indicate the usefulness of Google Trends index and encompassing tests for linear combination methods because linear combination methods coupled with CSET outperformed that coupled with CAGM and CUGM.

With Google Trends Web search index, the tourism sector may benefit from the use of linear combinations of constituents that survive encompassing tests to formulate business strategies for tourist destinations. A good forecasting practice by estimating ex ante forecasts post-COVID-19 can be further provided by scenario forecasting.

To improve the accuracy of combination forecasting, this research verifies the correlation between Google Trends index and combined forecasts in tourism along with encompassing tests.

過去的研究顯示 Google 搜尋趨勢資料有助於改善旅遊需求預測的準確度，本研究就此進一步探討 Google 搜尋趨勢網頁搜尋指標與涵蓋性檢定的使用對於組合預測準確度所造成的影響。

本研究以 Google 搜尋趨勢指標做為多變量灰色預測模式的解釋變數，並以單變量與多變量灰色模式產生各別預測值。在分別產生由所有的單變量模式 (CUGM)所有的模式 (CAGM), 以及經過涵蓋性檢定所留存下來之模式 (CSET) 所組成之集合後，就各別的組合集以常用的組合方法產生預測值。

以台灣的四個熱搜旅遊城市的旅遊人數進行三個組合集的預測準確度分析。涵蓋性檢定顯示多變量灰色模式在組合預測中扮演重要的角色，而結果亦呈現線性組合方法在 CSET優於在 CUGM 與 CAGM 的準確度，突顯搜尋趨勢指標與涵蓋性檢定對於線性組合方法的有用性。

藉由 Google 搜尋趨勢網頁搜尋指標與涵蓋性檢定，旅遊部門應可透過線性組合方法的預測規劃旅遊目的地的經營策略。新冠疫情下於各季的事前預測亦可結合情境預測具體呈現。

為提升組合預測在旅遊需求的預測準確度，本研究結合涵蓋性檢定以分析 Google 搜尋趨勢指標與組合預測準確度之間的關聯性。

旅遊需求,涵蓋性檢定,Google 搜尋趨勢,灰色預測,組合預測

研究型论文

Dado que el uso de los datos de Google Trends es útil para mejorar la precisión de las predicciones, este estudio examina si el uso del índice de búsqueda web de Google Trends combinado con la agregación de relevancia puede mejorar la precisión del predictor.

El modelo predictivo gris genera predicciones bajo un único modelo, mientras que el modelomultivariado utiliza el indicador Google Trends como variable explicativa. Se generaron tresensamblajes generales, incluido el Modelo armónico único (CUGM), los ensamblajes de todos loscomponentes (CAGM) y la prueba de presencia de componentes con predicción (CSET). Laspredicciones individuales encada grupo luego se combinan utilizando métodos de correlación deuso común.

Utilizando el número de turistas en las cuatro ciudades más investigadas de Taiwán, los tresgrupos combinados se clasificaron según su precisión. Las pruebas incluidas muestran que losmodelos multivariados en escala de grises son importantes para la predicción. Además, losresultados de las pruebas muestran que el índice de Google Trends y las pruebas que incluyenmétodos de suma lineal son útiles porque los métodos combinados con CSET funcionan majorque los métodos combinados con CSET. CAGM y VCUG.

La industria de viajes puede usar el índice de búsqueda web de Google Trends para desarrollarestrategias comerciales para atracciones basadas en un conjunto lineal de componentes.

Con el objetivo de mejorar la precisión de los pronósticos agregados, este estudio investiga larelación entre el índice de tendencias de Google y las expectativas generales de viaje junto con laevidencia global.

Demanda de viajes, Experiencia global, Tendencias de Google, Predicción gris

Trabajo de investigación

This research examines the extent to which the number of sales forecasting methods used by a company affects forecast accuracy and the extent to which organisation affects the number of sales forecast methods chosen. The objective is to better understand marketing management practices in this respect. Contextually, the study is part of the shift in sales forecasting research away from studies of accuracy per se to studies of organisation and implementation issues. It is widely recognised that objective techniques improve forecast accuracy, especially in the long run; yet, there is considerable evidence that such techniques are not widely used. The question of why there is such a discrepancy between practice and conventional wisdom, accounts, in large part, for this interest in organisation and implementation and the development of forecast models that incorporate implementation strategies.

Forecasting is an integral part of all business planning, and forecasting the outlook for housing is of interest to many firms in the housing construction sector. This research measures the performance of a number of industry forecasting bodies; this is done to provide users with an indicator of the value of housing forecasting undertaken in Australia. The accuracy of housing commencement forecasts of three Australian organisations – the Housing Industry Association (HIA), the Indicative Planning Council for the Housing Industry (IPC) and BIS‐Shrapnel – is examined through the empirical analysis of their published forecasts supplemented by qualitative data in the form of opinions elicited from several industry “experts” employed in these organisations. Forecasting performance was determined by comparing the housing commencement forecast with the actual data collected by the Australian Bureau of Statistics on an ex‐post basis. Although the forecasts cover different time periods, the level of accuracy is similar, at around 11‐13 per cent for four‐quarter‐ahead forecasts. In addition, national forecasts are more accurate than forecasts for individual states. This is the first research that has investigated the accuracy of both private and public sector forecasting of housing construction in Australia. This allows users of the information to better understand the performance of various forecasting organisations.