scholarly journals Model-Free Time-Aggregated Predictions for Econometric Datasets

Forecasting ◽  
2021 ◽  
Vol 3 (4) ◽  
pp. 920-933
Author(s):  
Kejin Wu ◽  
Sayar Karmakar
Keyword(s):  
Data Analysis ◽  
High Performance ◽  
Prediction Performance ◽  
Financial Data ◽  
New Method ◽  
Model Free ◽  
Short And Long Term ◽  
Forecasting Volatility ◽  
Long Term Prediction

Forecasting volatility from econometric datasets is a crucial task in finance. To acquire meaningful volatility predictions, various methods were built upon GARCH-type models, but these classical techniques suffer from instability of short and volatile data. Recently, a novel existing normalizing and variance-stabilizing (NoVaS) method for predicting squared log-returns of financial data was proposed. This model-free method has been shown to possess more accurate and stable prediction performance than GARCH-type methods. However, whether this method can sustain this high performance for long-term prediction is still in doubt. In this article, we firstly explore the robustness of the existing NoVaS method for long-term time-aggregated predictions. Then, we develop a more parsimonious variant of the existing method. With systematic justification and extensive data analysis, our new method shows better performance than current NoVaS and standard GARCH(1,1) methods on both short- and long-term time-aggregated predictions. The success of our new method is remarkable since efficient predictions with short and volatile data always carry great importance. Additionally, this article opens potential avenues where one can design a model-free prediction structure to meet specific needs.

scholarly journals Novel robust time series analysis for long-term and short-term prediction

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Hiroshi Okamura ◽  
Yutaka Osada ◽  
Shota Nishijima ◽  
Shinto Eguchi
Keyword(s):  
Robust Regression ◽  
Simulated Data ◽  
New Method ◽  
Short Term ◽  
Least Absolute Deviations ◽  
Term Prediction ◽  
Regression Approach ◽  
Short Term Prediction ◽  
Long Term Prediction

AbstractNonlinear phenomena are universal in ecology. However, their inference and prediction are generally difficult because of autocorrelation and outliers. A traditional least squares method for parameter estimation is capable of improving short-term prediction by estimating autocorrelation, whereas it has weakness to outliers and consequently worse long-term prediction. In contrast, a traditional robust regression approach, such as the least absolute deviations method, alleviates the influence of outliers and has potentially better long-term prediction, whereas it makes accurately estimating autocorrelation difficult and possibly leads to worse short-term prediction. We propose a new robust regression approach that estimates autocorrelation accurately and reduces the influence of outliers. We then compare the new method with the conventional least squares and least absolute deviations methods by using simulated data and real ecological data. Simulations and analysis of real data demonstrate that the new method generally has better long-term and short-term prediction ability for nonlinear estimation problems using spawner–recruitment data. The new method provides nearly unbiased autocorrelation even for highly contaminated simulated data with extreme outliers, whereas other methods fail to estimate autocorrelation accurately.

scholarly journals Serotonin Differentially Regulates Short- and Long-Term Prediction of Rewards in the Ventral and Dorsal Striatum

PLoS ONE ◽  
2007 ◽  
Vol 2 (12) ◽  
pp. e1333 ◽  
Author(s):  
Saori C. Tanaka ◽  
Nicolas Schweighofer ◽  
Shuji Asahi ◽  
Kazuhiro Shishida ◽  
Yasumasa Okamoto ◽  
...  
Keyword(s):  
Dorsal Striatum ◽  
Term Prediction ◽  
Short And Long Term ◽  
Long Term Prediction

Auscultation d'une culée du pont de Portneuf à l'aide d'une inclusion de béton instrumentée (CIUS)

1996 ◽  
Vol 23 (5) ◽  
pp. 1129-1136
Author(s):  
Axel-Pierre Bois ◽  
Mohamed Lachemi ◽  
Gérard Ballivy
Keyword(s):  
High Performance ◽  
High Performance Concrete ◽  
Coefficient Of Thermal Expansion ◽  
Concrete Bridge ◽  
Thermal Gradients ◽  
Bridge Monitoring ◽  
Shrinkage And Creep ◽  
First Results ◽  
Short And Long Term

The Portneuf Bridge, built in 1992, is the first air-entrained high-performance concrete bridge in North America. To understand its short and long term behaviour, an auscultation program has been set. Hence, a cylindrical concrete inclusion of the Université de Sherbrooke was installed in one of the abutments of the bridge. The aim of this study is to present the first results thus acquired. The analysis of the results allowed to calculate the coefficient of thermal expansion of the concrete and to assess deformation variations due to shrinkage and creep and the effects of rebar–concrete interaction in the upper abutment region. Moreover, the presence of thermal gradients, which creates nonisotropic deformations, has been established. Key words: high-performance concrete, deformations, thermal gradients, instrumentation, bridge, monitoring. [Journal translation]

O4-04-05: Differences in short- and long-term prediction of dementia in a 15-year longitudinal population study

2006 ◽  
Vol 2 ◽  
pp. S83-S83
Author(s):  
Simona F. Sacuiu ◽  
Boo Johansson ◽  
Svante Östling ◽  
Deborah Gustafson ◽  
Ingmar Skoog
Keyword(s):  
Population Study ◽  
Term Prediction ◽  
Short And Long Term ◽  
Long Term Prediction

scholarly journals Short and Long-Term Properties of High-Performance Concrete Containing Silica Fume for Bridge Deck Overlay

2005 ◽  
Vol 17 (5) ◽  
pp. 743-750
Author(s):  
Jong-Pil Won ◽  
Jung-Min Seo ◽  
Chang-Soo Lee ◽  
Hae-Kyun Park ◽  
Myeong-Sub Lee
Keyword(s):  
Silica Fume ◽  
High Performance ◽  
Bridge Deck ◽  
High Performance Concrete ◽  
Short And Long Term ◽  
Bridge Deck Overlay

Nonlinear Hierarchical MPC for Maximizing Aircraft Thermal Endurance

2020 ◽  
Author(s):  
Daniel D. Leister ◽  
Justin P. Koeln
Keyword(s):  
High Performance ◽  
Linear Models ◽  
Critical Role ◽  
Effective Control ◽  
Thermal Management System ◽  
Short And Long Term ◽  
Upper Level ◽  
Control Of Linear Systems ◽  
High Computational Efficiency

Abstract In modern high-performance aircraft, the Fuel Thermal Management System (FTMS) plays a critical role in the overall thermal energy management of the aircraft. Actuator and state constraints in the FTMS limit the thermal endurance and capabilities of the aircraft. Thus, an effective control strategy must plan and execute optimized transient fuel mass and temperature trajectories subject to these constraints over the entire course of operation. For the control of linear systems, hierarchical Model Predictive Control (MPC) has shown to be an effective approach to coordinating both short- and long-term system operation with reduced computational complexity. However, for controlling nonlinear systems, common approaches to system linearization may no longer be effective due to the long prediction horizons of upper-level controllers. This paper explores the limitations of using linear models for hierarchical MPC of the nonlinear FTMS found in aircraft. Numerical simulation results show that linearized models work well for lower-level controllers with short prediction horizons but lead to significant reductions in aircraft thermal endurance when used for upper-level controllers with long prediction horizons. Therefore, a mixed-linearity hierarchical MPC formulation is presented with a nonlinear upper-level controller and a linear lower-level controller to achieve both high performance and high computational efficiency.

scholarly journals DERN: Deep Ensemble Learning Model for Short- and Long-Term Prediction of Baltic Dry Index

2020 ◽  
Vol 10 (4) ◽  
pp. 1504 ◽  
Author(s):  
Imam Mustafa Kamal ◽  
Hyerim Bae ◽  
Sim Sunghyun ◽  
Heesung Yun
Keyword(s):  
Deep Learning ◽  
Network Models ◽  
Neural Network Models ◽  
Business Decisions ◽  
Term Prediction ◽  
Short And Long Term ◽  
The Baltic ◽  
Long Term Prediction ◽  
Baltic Dry Index

The Baltic Dry Index (BDI) is a commonly utilized indicator of global shipping and trade activity. It influences stakeholders’ and ship-owners’ decisions respecting investments, chartering, operational plans, and export and import activities. Accurate prediction of the BDI is very challenging due to its volatility, non-stationarity, and complexity. To help stakeholders and ship-owners make sound short- and long-term maritime business decisions and avoid market risk, we performed short- and long-term predictions of BDI using an ensemble deep-learning approach. In this study, we propose to apply recurrent neural network models for BDI prediction. The state-of-the-art of sequential deep-learning models such as RNN, LSTM, and GRU are employed to predict one- and multi-step-ahead BDI values. In order to increase the accuracy, we assemble the models. In experiments, we compared our results with those of traditional methods such as ARIMA and MLP. The results showed that our proposed method outperforms ARIMA, MLP, RNN, LSTM, and GRU in both short- and long-term prediction of BDI.

scholarly journals INTRODUCTION TO TIME-VARYING MODELING WITH MACROECONOMIC AND FINANCIAL DATA

2012 ◽  
Vol 16 (S2) ◽  
pp. 167-175 ◽  
Author(s):  
Fredj Jawadi
Keyword(s):  
Time Series ◽  
Time Variation ◽  
Moving Average ◽  
Statistical Properties ◽  
Financial Data ◽  
Time Varying ◽  
Arch Model ◽  
Financial Variables ◽  
Short And Long Term

The dynamics of macroeconomic and financial series has evolved swiftly and asymmetrically since the end of the 1970s, and their statistical properties have also changed over time, suggesting complex relationships between economic and financial variables. The transformations can be explained by considerable changes in householder's behavior, market structures, and economic systems and by the alternation of exogenous shocks and financial crises that have affected the economic cycle, with significant evidence of time variation in the major economic variables. Hence, there is a need for new econometric protocols to take such changes into consideration. The introduction of ARMA (autoregressive moving average models) by Box and Jenkins (1970) led to the development of time-series econometrics, which had a major impact on the conceptual analysis of economic and financial data. This type of modeling offered a transition from a static setup to a new modeling process that reproduces the time-varying features of macroeconomic and financial series. However, the ARMA modeling system retains the constancy of the first and second moments, limits the phases of a cycle to symmetrical instances, and only reproduces the dynamics of stationary variables. It thus fails to adequately reproduce the nonstationary relationships between major economic and financial variables. Abrupt changes in economies and financial systems have given evidence of nonstationary series whose statistical properties are also time-varying, making it necessary to develop new econometric tools to capture the time variation of economic and financial series in the mean and in the variance, and to apprehend their dynamics in the short and long term. Among the most important and influential studies in the 1980s' econometrics literature were therefore those that dealt with the introduction of the ARCH (autoregressive conditional heteroskedasticity) model by Engle (1982) and the cointegration theory by Engle and Granger (1987). The ARCH model, which focuses on the time-varying features of volatility structure, was a major breakthrough, as it highlighted the importance of the second moment of time series, while the cointegration framework enabled the short- and long-term dynamics of nonstationary variables to be modeled.

Sign in / Sign up

Export Citation Format

Share Document