Selecting Forecasting Methods

Armstrong, J. Scott

doi:10.2139/ssrn.1941247

Cited by 9 publications

(7 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These measures can be used to examine the size of forecast errors in both relative (percentage) and absolute (volume) terms. It is important to use more than one measure of errors, as no single measure has been shown to provide an unambiguous indication of forecast accuracy (Armstrong 2001b;Mathews and Diamantopoulos 1986).…”

Section: Methodsmentioning

confidence: 99%

Ex Ante Tourism Forecasting Assessment

Liu

Lin

et al. 2020

Journal of Travel Research

View full text Add to dashboard Cite

Although numerous studies have focused on forecasting international tourism demand, minimal light has been shed on the factors influencing the accuracy of real-world ex ante forecasting. This study evaluates the forecasting errors across various prediction horizons by analyzing the annually published forecasts of the Pacific Asia Tourism Association (PATA) from 2013 to 2017, comprising 765 origin–destination pairs covering 31 destinations in the region. The regression analysis shows that the variation in tourism demand and gross domestic product (GDP), covariation between tourism demand and GDP, order of lagged variables, origin, destination, and forecasting method all have significant effects on the forecasting accuracy over different horizons. This suggests that tourism forecasting should account for these factors in the future.

show abstract

Section: Methodsmentioning

confidence: 99%

Ex Ante Tourism Forecasting Assessment

Liu

Lin

et al. 2020

Journal of Travel Research

View full text Add to dashboard Cite

show abstract

“…Finally, the third category of methods for predicting market adoption, data analysis, also suffers from crucial difficulties when applied to major innovations (Armstrong, 2001b). Because of the shift in customer behaviour, past data regarding the diffusion of predecessors of the innovation can no longer be extrapolated to indicate the future adoption of such a major innovation.…”

Section: Existing Methodsmentioning

confidence: 99%

Imitation analysis

Langley

Pals

Ortt³

et al. 2009

Euro Jrnl of Inn Mnagmnt

View full text Add to dashboard Cite

PurposeThe purpose of this paper is to describe a method of estimating the likelihood that a person with particular characteristics will imitate a particular new behaviour (i.e. the use of an innovation). This estimation can be used to provide a new form of forecast for the likely market demand for an innovation.Design/methodology/approachThis method, termed imitation analysis, is based on imitation theories from the behavioural sciences and is applied in two recent case studies in The Netherlands: broadcast TV on mobile phones and a mobile friend‐network service.FindingsThese cases illustrate how: the market segments with the highest potential can be identified; marketing communication can be focused on specific issues important for each segment (e.g. based on the highest imitation potential); product design can be improved (by highlighting the characteristics with the most room for improving the imitation potential); and market demand can be modelled (the overall chance of imitation occurring).Practical implicationsManagement implications for the two services, as well as the usefulness of imitation analysis in forecasting studies, are discussed.Originality/valueThe paper expands on original work published in this journal in 2005, showing the value of the approach in real‐world settings.

show abstract

“…Different predictive models whose implementation can impact planning processes are used [35]. The processes used for the selection of the forecasting methods are described in [36]. In processing time-series data, one of the most commonly used methods includes data classification.…”

Section: Predictive Modelsmentioning

confidence: 99%

“…Several measures are used to determine the accuracy of conducted forecasts. Statistical criteria can be selected after deciding on the general type of forecasting method [36]. There are mainly four types of forecast-error metrics: scale-dependent, relativeerror, and scale-free error metrics.…”

Section: B) Forecasting Processmentioning

confidence: 99%

Distribution and Allocation of Network Resources Based on Predictive Analyses of Time-Series Telecommunications Data

Križanović

Vlaovic

Žagar

et al. 2022

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

With the fast development of different communication technologies, applications, and services, the adoption of advanced sensory and computing solutions, such as the various Internet of Things (IoT) and mobile computing solutions, is continuously growing. The massive adoption of mobile computing and IoT sensory devices encouraged the continuous growth of generated network traffic. Therefore, the selection of adequate solutions for efficient data processing became necessary. Despite numerous advantages arising from effective data processing, operators and enterprises working within the ICT domain have only limited amounts of available networking resources to store, process, and use valuable information extracted from large quantities of gathered data. In this paper, an optimal planning process and prediction of usage of network resources is examined. It takes into consideration the results of predictive modeling processes based on available sets of time series telecommunications data. The given forecasts enable effective selection of network architectures, as well as the distribution and allocation of network resources considering the cloud, edge, and fog networking concepts.

show abstract

Selecting Forecasting Methods

Cited by 9 publications

References 60 publications

Ex Ante Tourism Forecasting Assessment

Ex Ante Tourism Forecasting Assessment

Imitation analysis

Distribution and Allocation of Network Resources Based on Predictive Analyses of Time-Series Telecommunications Data

Contact Info

Product

Resources

About