Search results
1 – 10 of over 34000This paper investigates how various strategies for combining forecasts, both simple and optimised approaches, are compared with popular individual risk models in estimating…
Abstract
Purpose
This paper investigates how various strategies for combining forecasts, both simple and optimised approaches, are compared with popular individual risk models in estimating value-at-risk (VaR) and expected shortfall (ES) in emerging market at alternative risk levels.
Design/methodology/approach
Using the case study of the Vietnamese stock market, the author produced one-day-ahead VaR and ES forecast from seven individual risk models and ten alternative forecast combinations. Next, the author employed a battery of backtesting procedures and alternative loss functions to evaluate the global predictive accuracy of the different methods. Finally, the author investigated the relative performance over time of VaR and ES forecasts using fluctuation test.
Findings
The empirical results indicate that, although combined forecasts have reasonable predictive abilities, they are often outperformed by one individual risk model. Furthermore, the author showed that the complex combining methods with optimised weighting functions do not perform better than simple combining methods. The fluctuation test suggests that the poor performance of combined forecasts is mainly due to their inability to cope with periods of instability.
Research limitations/implications
This study reveals the limitation of combining strategies in the one-day-ahead VaR and ES forecasts in emerging markets. A possible direction for further research is to investigate whether this finding holds for multi-day ahead forecasts. Moreover, the inferior performance of combined forecasts during periods of instability motivates further research on the combining strategies that take into account for potential structure breaks in the performance of individual risk models. A potential approach is to improve the individual risk models with macroeconomic variables using a mixed-data sampling approach.
Originality/value
First, the authors contribute to the literature on the forecasting combinations for VaR and ES measures. Second, the author explored a wide range of alternative risk models to forecast both VaR and ES with recent data including periods of the COVID-19 pandemic. Although forecast combination strategies have been providing several good results in several fields, the literature of forecast combination in the VaR and ES context is surprisingly limited, especially for emerging market returns. To the best of the author’s knowledge, this is the first study investigating predictive power of combining methods for VaR and ES in an emerging market.
Details
Keywords
Patrick J. Wilson and John Okunev
Over the last decade or so there has been an increased interest in combining the forecasts from different models. Pooling the forecast outcomes from different models has been…
Abstract
Over the last decade or so there has been an increased interest in combining the forecasts from different models. Pooling the forecast outcomes from different models has been shown to improve out‐of‐sample forecast test statistics beyond any of the individual component techniques. The discussion and practice of forecast combination has revolved around the pooling of results from individual forecasting methodologies. A different approach to forecast combination is followed in this paper. A method is used in which negatively correlated forecasts are combined to see if this offers improved out‐of‐sample forecasting performance in property markets. This is compared with the outcome from both the original model and with benchmark naïve forecasts over three 12‐month out‐of‐sample periods. The study will look at securitised property in three international property markets – the USA, the UK and Australia.
Details
Keywords
Xunfa Lu, Kang Sheng and Zhengjun Zhang
This paper aims to better jointly estimate Value at Risk (VaR) and expected shortfall (ES) by using the joint regression combined forecasting (JRCF) model.
Abstract
Purpose
This paper aims to better jointly estimate Value at Risk (VaR) and expected shortfall (ES) by using the joint regression combined forecasting (JRCF) model.
Design/methodology/approach
Combining different forecasting models in financial risk measurement can improve their prediction accuracy by integrating the individual models’ information. This paper applies the JRCF model to measure VaR and ES at 5%, 2.5% and 1% probability levels in the Chinese stock market. While ES is not elicitable on its own, the joint elicitability property of VaR and ES is established by the joint consistent scoring functions, which further refines the ES’s backtest. In addition, a variety of backtesting and evaluation methods are used to analyze and compare the alternative risk measurement models.
Findings
The empirical results show that the JRCF model outperforms the competing models. Based on the evaluation results of the joint scoring functions, the proposed model obtains the minimum scoring function value compared to the individual forecasting models and the average combined forecasting model overall. Moreover, Murphy diagrams’ results further reveal that this model has consistent comparative advantages among all considered models.
Originality/value
The JRCF model of risk measures is proposed, and the application of the joint scoring functions of VaR and ES is expanded. Additionally, this paper comprehensively backtests and evaluates the competing risk models and examines the characteristics of Chinese financial market risks.
Details
Keywords
Chen‐Chun Lin, Ying‐Hwa Tang, Joseph Z. Shyu and Yi‐Ming Li
The purpose of this paper is to propose an approach to achieve better accuracy in technology forecasting (TF) by providing the concepts of the service components and service…
Abstract
Purpose
The purpose of this paper is to propose an approach to achieve better accuracy in technology forecasting (TF) by providing the concepts of the service components and service composition based on the theory of the combining forecasts. Next, it adopts three quantitative analyses as service components to form service composition. This will support the need of more predictable TF, which raises the accuracy of the quantitative analysis and, at the same time, presents the service composition logic in a consistent manner in the form of customized TF.
Design/methodology/approach
This paper provides a systematic analysis of the technology forecasts for third‐generation (3G) telecommunication industry. This systematic approach mainly unifies the Bass model, logit model, and least squares analysis forecasting techniques, along with a reasonable assessment of the scope for the normal curve (±1 standard deviation), and attempts to find the maximum possibility frontier of the predictive value.
Findings
Through the integration and comparison of these three techniques, not only can the predicted values of the three forecasting methods be determined, but a preferred solution can also be derived through new methods, and in return, to investigate better accuracy and performances. Such an approach can also integrate the advantages of various methods to provide a prediction interval, as well as objective and realistic projections.
Research limitations/implications
This envisaged concept of “service component and service composition” is an integration of backing up in TF instruments in selection and reselection, which in return, provide optimization of service composition and accuracy maximization, as well as better performance prediction. A well‐known limitation of this research is that sudden technology breakthroughs are often unforeseeable in the majority of main‐stream quantitative analyses.
Originality/value
Constructing a new effective approach as results of “service component and service composition” can be compared to the traditional research methods such as Delphi method or other mathematical algorithms. This method generally produces higher quality forecasts than those attained from a single source.
Details
Keywords
Past research has shown that forecast combination typically improves demand forecast accuracy even when only two component forecasts are used; however, systematic bias in the…
Abstract
Past research has shown that forecast combination typically improves demand forecast accuracy even when only two component forecasts are used; however, systematic bias in the component forecasts can reduce the effectiveness of combination. This study proposes a methodology for combining demand forecasts that are biased. Data from an actual manufacturing shop are used to develop the methodology and compare its accuracy with the accuracy of the standard approach of correcting for bias prior to combination. Results indicate that the proposed methodology outperforms the standard approach.
Details
Keywords
Nada R. Sanders and Larry P. Ritzman
Accurate forecasting has become a challenge for companies operating in today's business environment, characterized by high uncertainty and short response times. Rapid…
Abstract
Accurate forecasting has become a challenge for companies operating in today's business environment, characterized by high uncertainty and short response times. Rapid technological innovations and e‐commerce have created an environment where historical data are often of limited value in predicting the future. In business organizations, the marketing function typically generates sales forecasts based on judgmental methods that rely heavily on subjective assessments and “soft” information, while operations rely more on quantitative data. Forecast generation rarely involves the pooling of information from these two functions. Increasingly, successful forecasting warrants the use of composite methodologies that incorporate a range of information from traditional quantitative computations usually used by operations, to marketing's judgmental assessments of markets. The purpose of this paper is to develop a framework for the integration of marketing's judgmental forecasts with traditional quantitative forecasting methods. Four integration methodologies are presented and evaluated relative to their appropriateness in combining forecasts within an organizational context. Our assessment considers human factors such as ownership, and the location of final forecast generation within the organization. Although each methodology has its strengths and weaknesses, not every methodology is appropriate for every organizational context.
Details
Keywords
Kenneth D. Lawrence, Dinesh R. Pai and Sheila M. Lawrence
This chapter proposes a fuzzy approach to forecasting using a financial data set. The methodology used is multiple objective linear programming (MOLP). Selecting an individual…
Abstract
This chapter proposes a fuzzy approach to forecasting using a financial data set. The methodology used is multiple objective linear programming (MOLP). Selecting an individual forecast based on a single objective may not make the best use of available information for a variety of reasons. Combined forecasts may provide a better fit with respect to a single objective than any individual forecast. We incorporate soft constraints and preemptive additive weights into a mathematical programming approach to improve our forecasting accuracy. We compare the results of our approach with the preemptive MOLP approach. A financial example is used to illustrate the efficacy of the proposed forecasting methodology.
Muhammad AsadUllah, Muhammad Adnan Bashir and Abdur Rahman Aleemi
The purpose of this study is to examine the accuracy of combined models with the individual models in terms of forecasting Euro against US dollar during COVID-19 era. During…
Abstract
Purpose
The purpose of this study is to examine the accuracy of combined models with the individual models in terms of forecasting Euro against US dollar during COVID-19 era. During COVID, the euro shows sharp fluctuation in upward and downward trend; therefore, this study is keen to find out the best-fitted model which forecasts more accurately during the pandemic.
Design/methodology/approach
The descriptive design has been adopted in this research. The three univariate models, i.e. autoregressive integrated moving averages (ARIMA), Naïve, exponential smoothing (ES) model, and one multivariate model, i.e. nonlinear autoregressive distributive lags (NARDL), are selected to forecast the exchange rate of Euro against the US dollar during the COVID. The above models are combined via equal weights and var-cor methods to find out the accuracy of forecasting as Poon and Granger (2003) showed that combined models can forecast better than individual models.
Findings
NARDL outperforms all remaining individual models, i.e. ARIMA, Naïve and ES. By applying a combination of different models via different techniques, the combination of NARDL and Naïve models outperforms all combination of models by scoring the least mean absolute percentage error value, i.e. 1.588. The combined forecasting of NARDL and Naïve techniques under var-cor method also outperforms the forecasting accuracy of individual models other than NARDL. It means the euro exchange rate against the US dollar which is dependent upon the macroeconomic fundamentals and recent observations of the time series.
Practical implications
The findings could help the FOREX market, hedgers, traders, businessmen, policymakers, economists, financial managers, etc., to minimize the risk indulged in global trade. It also helps to produce more accurate results in different financial models, i.e. capital asset pricing model and arbitrage pricing theory, because their findings may not be useful if exchange rate fluctuations do not trace effectively.
Originality/value
The NARDL models have been applied previously in different time series and only limited to the asymmetric or symmetric relationships. This study is using it for the forecasting exchange rate which is almost abandoned in earlier literature. Furthermore, this study combined the NARDL with univariate models to produce the accuracy which itself is a novelty. Moreover, the findings help to enhance the effectiveness of different financial theories as well.
Details
Keywords
Benedict von Ahlefeldt-Dehn, Marcelo Cajias and Wolfgang Schäfers
Commercial real estate and office rental values, in particular, have long been the focus of research. Several forecasting frameworks for office rental values in multivariate and…
Abstract
Purpose
Commercial real estate and office rental values, in particular, have long been the focus of research. Several forecasting frameworks for office rental values in multivariate and univariate fashions have been proposed. Recent developments in time series forecasting using machine learning and deep learning methods offer an opportunity to update traditional univariate forecasting frameworks.
Design/methodology/approach
With the aim to extend research on univariate rent forecasting a hybrid methodology combining both ARIMA and a neural network model is proposed to exploit the unique strengths of both methods in linear and nonlinear modelling. N-BEATS, a deep learning algorithm that has demonstrated state-of-the-art forecasting performance in major forecasting competitions, are explained. With the ARIMA model, it is jointly applied to the office rental dataset to produce forecasts for four-quarters ahead.
Findings
When the approach is applied to a dataset of 21 major European office cities, the results show that the ensemble model can be an effective approach to improve the prediction accuracy achieved by each of the models used separately.
Practical implications
Real estate forecasting is essential for assessing the value of managing portfolios and for evaluating investment strategies. The approach applied in this paper confirms the heterogeneity of real estate markets. The application of mixed modelling via linear and nonlinear methods decreases the uncertainty of abrupt changes in rents.
Originality/value
To the best of the authors' knowledge, no such application of a hybrid model updating classical statistical forecasting with a deep learning neural network approach in the field of commercial real estate rent forecasting has been undertaken.
Details