Search results

This study aims to make investment decisions in stock markets using forecasting-Markowitz based decision-making approaches.

The authors’ approach offers the use of time series prediction methods including autoregressive, autoregressive moving average and artificial neural network, rather than calculating the expected rate of return based on distribution.

The results show that using time series prediction methods has a significant effect on improving investment decisions and the performance of the investments.

In this study, in contrast to previous studies, the alteration in the Markowitz model started with the investment expected rate of return. For this purpose, instead of considering the distribution of returns and determining the expected returns, time series prediction methods were used to calculate the future return of each asset. Then, the results of different time series methods replaced the expected returns in the Markowitz model. Finally, the overall performance of the method, as well as the performance of each of the prediction methods used, was examined in relation to nine stock market indices.

The purpose of this paper is to present an improved particle filter-based attitude estimator for a quadrotor unmanned aerial vehicle (UAV) that addresses the degeneracy issues.

Control of a quadrotor is not sufficient enough without an estimator to eliminate the noise from low-cost sensors. In this work, particle filter-based attitude estimator is proposed and used for nonlinear quadrotor dynamics. But, since recursive Bayesian estimation steps may rise degeneracy issues, the proposed scheme is improved with four different and widely used resampling algorithms.

Robustness of the proposed schemes is tested under various scenarios that include different levels of uncertainty and different particle sizes. Statistical analyses are conducted to assess the error performance of the schemes. According to the statistical analysis, the proposed estimators are capable of reducing sensor noise up to 5x, increasing signal to noise ratio up to 2.5x and reducing the uncertainty bounds up to 36x with root mean square value of as low as 0.0024, mean absolute error value of 0.036, respectively.

To the best of the authors’ knowledge, the originality of this paper is to propose a robust particle filter-based attitude estimator to eliminate the low-cost sensor errors of quadrotor UAVs.

This paper aims to extend the research into company financial forecasts by modelling naïve earnings forecasts derived from normalised historic accounting data disclosed during Australian initial public offerings (IPOs). It seeks to investigate naïve forecast errors and compare them against their management forecast counterparts. It also seeks to investigate determinants of differential error behaviour.

IPOs were sampled and their prospectus forecasts, historic financial data and subsequent actual financial performance were analysed. Directional and absolute forecast error behaviour was analysed using univariate and multivariate techniques.

Systematic factors associated with error behaviour were observed across the management forecasts and the naïve forecasts, the most notable being audit quality. In certain circumstances, the naïve forecasts performed at least as well as management forecasts. In particular, forecast interval was an important discriminator for accuracy, with the superiority of management forecasts only observed for shorter forecast intervals.

The results imply a level of “disclosure management” regarding company IPO forecasts and normalised historic accounting data, with forecast overestimation and error size more extreme in the absence of higher quality third‐party monitoring services via the audit process. The results also raise questions regarding the serviceability of normalised historic financial information disclosed in prospectuses, in that many of those data do not appear to enhance the forecasting process, particularly when accompanied by published management forecasts and shorter forecast intervals.

This study evaluates the Federal Reserve (Fed) initial and final forecasts of the unemployment rate for 1983Q1-2018Q4. The Fed initial forecasts in a typical quarter are made in the first month (or immediately after), and the final forecasts are made in the third month of the quarter. The analysis also includes the private forecasts, which are made close to the end of the second month of the quarter.

In evaluating the multi-period forecasts, the study tests for systematic bias, directional accuracy, symmetric loss, equal forecast accuracy, encompassing and orthogonality. For every test equation, it employs the Newey–West procedure in order to obtain the standard errors corrected for both heteroscedasticity and inherent serial correlation.

Both Fed and private forecasts beat the naïve benchmark and predict directional change under symmetric loss. Fed final forecasts are more accurate than initial forecasts, meaning that predictive accuracy improves as more information becomes available. The private and Fed final forecasts contain distinct predictive information, but the latter produces significantly lower mean squared errors. The results are mixed when the study compares the private with the Fed initial forecasts. Additional results indicate that Fed (private) forecast errors are (are not) orthogonal to changes in consumer expectations about future unemployment. As such, consumer expectations can potentially help improve the accuracy of private forecasts.

Unlike many other studies, this study focuses on the unemployment rate, since it is an important indicator of the social cost of business cycles, and thus its forecasts are of special interest to policymakers, politicians and social scientists. Accurate unemployment rate forecasts, in particular, are essential for policymakers to design an optimal macroeconomic policy.

This study is concerned with evaluating the Federal Reserve forecasts of light motor vehicle sales. The goal is to assess accuracy gains from using consumer vehicle-buying attitudes and expectations about future business conditions derived from the long-running Michigan Surveys of Consumers.

Simplicity is a core principle in forecasting, and the literature provides plentiful evidence that combining forecasts from different methods and models reduces out-of-sample forecast errors if the methods and models are valid. As such, the authors construct a simple vector autoregressive (VAR) model that incorporates consumer vehicle-buying attitudes and expectations about future business conditions. Comparable forecasts of vehicle sales from this model are then combined with the Federal Reserve forecasts to assess accuracy gains.

The findings for 1994–2016 indicate that the Federal Reserve and VAR forecasts contain distinct and useful predictive information, and the combination of the two forecasts shows reductions in forecast errors that are more significant at longer horizons. The authors thus conclude that there are accuracy gains from using consumer survey responses.

This is the first study that is concerned with evaluating the Federal Reserve forecasts of vehicle sales and examines whether there are accuracy gains from using consumer vehicle-buying attitudes and expectations.

The emphasis on inter-organizational systems gave rise to concerns about inter-organizational relationships as trading partners became aware of the socio-political factors and trust that affect their relationships. This paper examines the importance of inter-organizational-trust in business-to-business E-commerce organizations. It examines how inter-organizational relationships impact trading partner trust, perceived benefits, perceived risks, and technology trust mechanisms in E-commerce that can in turn influence outcomes of business-to-business E-commerce. This paper develops a conceptual model and tests the model using a case study research methodology. The aim is to solicit qualitative in depth understanding of inter-organizational-trust in business-to-business E-commerce. Eight organizations from a cross section of industries that formed four bi-directional dyads participated in the third stage of this study. The first two stages include exploratory case studies in three organizations in the automotive industry that applied EDI via Value-Added-Networks in 1997, and a nationwide survey of organizations that examined the extent of E-commerce adoption in Australia and New Zealand in 1998. The findings identify the need for trustworthy business relationships in an E-commerce environment.

The purpose of this study is to develop and test a new deep learning model to predict aircraft fuel consumption. For this purpose, real data obtained from different landings and take-offs were used. As a result, a new hybrid convolutional neural network (CNN)-bi-directional long short term memory (BiLSTM) model was developed as intended.

The data used are divided into training and testing according to the k-fold 5 value. In this study, 13 different parameters were used together as input parameters. Fuel consumption was used as the output parameter. Thus, the effect of many input parameters on fuel flow was modeled simultaneously using the deep learning method in this study. In addition, the developed hybrid model was compared with the existing deep learning models long short term memory (LSTM) and BiLSTM.

In this study, when tested with LSTM, one of the existing deep learning models, values of 0.9162, 6.476, and 5.76 were obtained for R², root mean square error (RMSE), and mean absolute percentage error (MAPE), respectively. For the BiLSTM model when tested, values of 0.9471, 5.847 and 4.62 were obtained for R², RMSE and MAPE, respectively. In the proposed hybrid model when tested, values of 0.9743, 2.539 and 1.62 were obtained for R², RMSE and MAPE, respectively. The results obtained according to the LSTM and BiLSTM models are much closer to the actual fuel consumption values. The error of the models used was verified against the actual fuel flow reports, and an average absolute percent error value of less than 2% was obtained.

In this study, a new hybrid CNN-BiLSTM model is proposed. The proposed model is trained and tested with real flight data for fuel consumption estimation. As a result of the test, it is seen that it gives much better results than the LSTM and BiLSTM methods found in the literature. For this reason, it can be used in many different engine types and applications in different fields, especially the turboprop engine used in the study. Because it can be applied to different engines than the engine type used in the study, it can be easily integrated into many simulation models.

The present paper aims to assess the reliability and the limitations of analysing flight stability of a box-wing aircraft configuration known as PrandtlPlane by means of methods conceived for conventional aircraft and well known in the literature.

Results obtained by applying vortex lattice methods to PrandtlPlane configuration, validated previously with wind tunnel tests, are compared to the output of a “Roskam-like” method, here defined to model the PrandtlPlane features.

The comparisons have shown that the “Roskam-like” model gives accurate predictions for both the longitudinal stability margin and dihedral effect, whereas the directional stability is always overestimated.

The method here proposed and related achievements are valid only for subsonic conditions. The poor reliability related to lateral-directional derivatives estimations may be improved implementing different models known from the literature.

The possibility of applying a faster method as the “Roskam-like” one here presented has two main implications: it allows to implement faster analyses in the conceptual and preliminary design of PrandtlPlane, providing also a tool for the definition of the design space in case of optimization approaches and it allows to implement a scaling procedure, to study families of PrandtlPlanes or different aircraft categories.

This paper is part of the activities carried out during the PARSIFAL project, which aims to demonstrate that the introduction of PrandtlPlane as air transport mean can fuel consumption and noise impact, providing a sustainable answer to the growing air passenger demand envisaged for the next decades.

The originality of this paper lies in the attempt of adopting analysis method conceived for conventional airplanes for the analysis of a novel configuration. The value of the work is represented by the knowledge concerning experimental results and design methods on the PrandtlPlane configuration, here made available to define a new analysis tool.

Property market models have the overriding aim of predicting reasonable estimates of key dependent variables (demand, supply, rent, yield, vacancy and net absorption rate). These can be based on independent drivers of core property and economic activities. Accurate predictions can only be conducted when ample quantitative data are available with fewer uncertainties. However, a broad-fronted social, technical and ecological evolution can throw up sudden, unexpected shocks that result in the econometric outputs sceptical to unknown risk factors. Therefore, the purpose of this paper is to evaluate Australian office market forecast accuracy and to determine whether the forecasts capture extreme downside risk events.

This study follows a quantitative research approach, using secondary data analysis to test the accuracy of economists’ forecasts. The forecast accuracy evaluation encompasses the measurement of economic and property forecasts under the following phases: testing for the forecast accuracy; analysing outliers of forecast errors; and testing of causal relationships. Forecast accuracy measurement incorporates scale independent metrics that include Theil’s U values (U1 and U2) and mean absolute scaled error. Inter-quartile range rule is used for the outlier analysis. To find the causal relationships among variables, the time series regression methodology is utilised, including multiple regression analysis and Granger causality developed under the vector auto regression (VAR).

The credibility of economic and property forecasts was questionable around the period of the Global Financial Crisis (GFC); a significant man-made Black Swan event. The forecast accuracy measurement highlighted rental movement and net absorption forecast errors as the critical inaccurate predictions. These key property variables are explained by historic information and independent economic variables. However, these do not explain the changes when error time series of the variables were concerned. According to VAR estimates, all property variables have a significant causality derived from the lagged values of Australian S&P/ASX 200 (ASX) forecast errors. Therefore, lagged ASX forecast errors could be used as a warning signal to adjust property forecasts.

Secondary data were obtained from the premier Australian property markets: Canberra, Sydney, Brisbane, Adelaide, Melbourne and Perth. A limited ten-year timeframe (2001-2011) was used in the ex-post analysis for the comparison of economic and property variables. Forecasts ceased from 2011, due to the discontinuity of the Australian Financial Review quarterly survey of economists; the main source of economic forecast data.

The research strongly recommended naïve forecasts for the property variables, as an input determinant in each office market forecast equation. Further, lagged forecast errors in the ASX could be used as a warning signal for the successive property forecast errors. Hence, data adjustments can be made to ensure the accuracy of the Australian office market forecasts.

The paper highlights the critical inaccuracy of the Australian office market forecasts around the GFC. In an environment of increasing incidence of unknown events, these types of risk events should not be dismissed as statistical outliers in real estate modelling. As a proactive strategy to improve office market forecasts, lagged ASX forecast errors could be used as a warning signal. This causality was mirrored in rental movements and total vacancy forecast errors. The close interdependency between rents and vacancy rates in the forecasting process and the volatility in rental cash flows reflects on direct property investment and subsequently on the ASX, is therefore justified.

Compares the predictive performance of artificial neural networks to hedonic pricing models, a more traditional valuation tool. The results document similar predictive performance evidenced from both techniques, which contradicts some of the earlier studies which support a position of artificial neural network superiority. Demonstrates that at least 18 per cent of the “normal” property predictions and over 70 per cent of the “outlier” property predictions contained valuation errors greater than 15 per cent of the actual sales price. The combination of these substantial errors and the model‐optimization costs incurred motivate a message of caution before artificial neural networks are adopted by the real estate valuation and/or lending industries.