In this paper, a novel decision support system using a computational efficient functional link artificial neural network (CEFLANN) and a set of rules is proposed to generate the trading decisions more effectively. Here the problem of stock trading decision prediction is articulated as a classification problem with three class values representing the buy, hold and sell signals. The CEFLANN network used in the decision support system produces a set of continuous trading signals within the range 0 to 1 by analyzing the nonlinear relationship exists between few popular technical indicators. Further the output trading signals are used to track the trend and to produce the trading decision based on that trend using some trading rules. The novelty of the approach is to engender the profitable stock trading decision points through integration of the learning ability of CEFLANN neural network with the technical analysis rules. For assessing the potential use of the proposed method, the model performance is also compared with some other machine learning techniques such as Support Vector Machine (SVM), Naive Bayesian model, K nearest neighbor model (KNN) and Decision Tree (DT) model.
The motivation behind this research is to innovatively combine new methods like wavelet, principal component analysis (PCA), and artificial neural network (ANN) approaches to analyze trade in today’s increasingly difficult and volatile financial futures markets. The main focus of this study is to facilitate forecasting by using an enhanced denoising process on market data, taken as a multivariate signal, in order to deduct the same noise from the open-high-low-close signal of a market. This research offers evidence on the predictive ability and the profitability of abnormal returns of a new hybrid forecasting model using Wavelet-PCA denoising and ANN (named WPCA-NN) on futures contracts of Hong Kong’s Hang Seng futures, Japan’s NIKKEI 225 futures, Singapore’s MSCI futures, South Korea’s KOSPI 200 futures, and Taiwan’s TAIEX futures from 2005 to 2014. Using a host of technical analysis indicators consisting of RSI, MACD, MACD Signal, Stochastic Fast %K, Stochastic Slow %K, Stochastic %D, and Ultimate Oscillator, empirical results show that the annual mean returns of WPCA-NN are more than the threshold buy-and-hold for the validation, test, and evaluation periods; this is inconsistent with the traditional random walk hypothesis, which insists that mechanical rules cannot outperform the threshold buy-and-hold. The findings, however, are consistent with literature that advocates technical analysis.
환전은 한 통화를 다른 통화에 대해 거래하는 것입니다. FOREX 비율은 많은 상관관계가 있는 경제적, 정치적, 심리적 요인의 영향을 받으므로 예측하는 것은 어려운 작업입니다. FOREX 비율을 예측하는 몇 가지 방법에는 통계 분석, 시계열 분석, 퍼지 시스템, 신경망 및 하이브리드 시스템이 있습니다. 이러한 방법은 교환을 정확하게 예측하는 문제가 있습니다. FOREX 시장의 미래 속도를 예측하기 위해 인공 신경망(ANN)과 하이브리드 신경 퍼지 시스템(ANFIS)이 제안됩니다. MLP는 환율의 상승 또는 하락을 예측하는 데 사용되며 ANFIS 모델은 다음 날의 환율을 예측하는 데 사용됩니다. 실험을 위해 외환 시장의 USDINR 환율이 사용됩니다. 평균 제곱 오차(MSE) 및 평균 절대 오차(MAE)는 성능 지표로 사용됩니다. ANN은 훈련 동안 0.033의 MSE와 0.0002의 MAE를 달성한 반면 ANFIS 모델은 0.024의 MSE와 6.7x10-8의 MAE를 달성했습니다. ANN은 0.003의 MSE와 0.00082의 MAE를 달성한 반면 ANFIS 모델은 테스트 단계에서 0.02의 MSE와 0.00792의 MAE를 달성했습니다.
지금까지 마케팅 문헌은 엄격한 기능적 형태의 매력 모델만을 고려했습니다. 하나의 숨겨진 레이어가 있는 퍼셉트론을 통해 브랜드의 매력 가치를 평가하는 도입된 신경망 기반 접근 방식으로 더 큰 유연성을 얻을 수 있습니다. 로그 비율 변환된 시장 점유율을 종속 변수로 사용하여 확률적 경사하강법과 준-뉴턴 방법을 사용하여 매개변수를 추정합니다. 매장 수준 데이터의 경우 신경망 모델이 더 잘 수행되며 잘 알려진 MNL 매력 모델과 질적으로 다른 가격 응답을 의미합니다. 이러한 경쟁 모델의 가격 탄력성은 또한 특정 경영상의 의미를 낳습니다. (저자의 초록)
This paper propose that the combination of smoothing approach taking into account the entropic information provided by Renyi method, has an acceptable performance in term of forecasting errors. The methodology of the proposed scheme is examined through benchmark chaotic time series, such as Mackay Glass, Lorenz, Henon maps, the Lynx and rainfall from Santa Francisca series, with addition of white noise by using neural networks-based energy associated (EAS) predictor filter modified by Renyi entropy of the series. In particular, when the time series is short or long, the underlying dynamical system is nonlinear and temporal dependencies span long time intervals, in which this are also called long memory process. In such cases, the inherent nonlinearity of neural networks models and a higher robustness to noise seem to partially explain their better prediction performance when entropic information is extracted from the series. Then, to demonstrate that permutation entropy is computationally efficient, robust to outliers, and effective to measure complexity of time series, computational results are evaluated against several non-linear ANN predictors proposed before to show the predictability of noisy rainfall and chaotic time series reported in the literature.
W e propose a forecasting procedure based on multivariate 금융 시장에서도. 데이터 압축 프로세스는 원래 금융 시계열을 시간 데이터 블록으로 재정의하여 여러 시간 간격의 시간 정보를 분석합니다. 알. 분석은 multiv를 통해 수행됩니다. 지원 벡터 회귀 내에서 동적 커널을 생성합니다. 여 e는 또한 정확성에 대한 희생 없이 계산적으로 효율적인 금융 시계열에 대한 두 개의 커널을 제안합니다. . 방법론의 효과는 도전적인 S&P500 시장 예측에 대한 실증적 실험에 의해 입증되었습니다.
This study presents a neural network & web-based decision support system (DSS) for foreign exchange (forex) forecasting and trading decision, which is adaptable to the needs of financial organizations and individual investors. In this study, we integrate the back-propagation neural network (BPNN)- based forex rolling forecasting system to accurately predict the change in direction of daily exchange rates, and the Web-based forex trading decision support system to obtain forecasting data and provide some investment decision suggestions for financial practitioners. This research reveals the structure of the DSS by the description of an integrated framework, and meantime we find that the DSS is integrated, user-oriented by its implementation, and practical applications reveal that this DSS demonstrates very high forecasting accuracy and its trading recommendations are reliable.
노이즈 주입은 신경망(NN)에서 과적합을 완화하기 위한 기성품 방법입니다. 드롭아웃 및 쉐이크아웃 절차에서 구현된 베르누이 노이즈 주입의 최근 개발은 깊은 신경망을 정규화하는 데 노이즈 주입의 효율성과 실행 가능성을 보여줍니다. 우리는 적응형 가우시안 잡음을 심층 신경망에 주입하여 새로운 정규화 기법인 화이트아웃을 제안합니다. Whiteout은 NN 훈련 중에 유연성을 제공하는 세 가지 조정 매개변수를 제공합니다. 우리는 폐쇄형 페널티 항이 있는 일반화 선형 모델의 맥락에서 화이트아웃이 결정론적 최적화 목적 함수와 연관되어 있으며 특수한 경우로 올가미, 능선 회귀, 적응형 올가미 및 탄성 망을 포함한다는 것을 보여줍니다. 우리는 또한 화이트아웃이 입력 및 은닉 노드에서 작고 미미한 섭동이 있는 상태에서 NN 모델의 강력한 학습으로 볼 수 있음을 보여줍니다. 드롭아웃에 비해 화이트아웃은 엘1 정규화. 셰이크아웃에 비해 화이트아웃의 벌점 목적 함수는 더 나은 수렴 동작을 가지며 주입된 노이즈의 연속성을 고려할 때 더 안정적입니다. 우리는 이론적으로 화이트아웃이 있는 노이즈 섭동 경험적 손실 함수가 이상적인 손실 함수에 거의 확실하게 수렴하고 전자 손실 함수를 최소화하여 얻은 NN 매개변수의 추정치가 이상적인 손실 함수를 최소화하여 얻은 것과 일치한다는 것을 이론적으로 설정합니다. 계산적으로 화이트아웃은 역전파 알고리즘에 통합될 수 있으며 계산적으로 효율적입니다. 분류에서 훈련 NN에서 탈락 및 쉐이크아웃에 대한 화이트아웃의 우월성은 MNIST 데이터를 사용하여 입증되었습니다.
깊이는 네트워크 성능을 향상시키는 경향이 있지만 더 깊은 네트워크는 더 비선형적인 경향이 있기 때문에 그래디언트 기반 훈련을 더 어렵게 만듭니다. 최근에 제안된 지식 증류 접근법은 작고 실행이 빠른 모델을 얻는 것을 목표로 하며 학생 네트워크가 더 큰 교사 네트워크 또는 네트워크 앙상블의 부드러운 출력을 모방할 수 있음을 보여주었습니다. 본 논문에서는 출력뿐만 아니라 교사가 학습한 중간 표현을 힌트로 사용하여 교사보다 더 깊고 얇은 학생의 훈련이 가능하도록 이 아이디어를 확장하여 훈련 과정과 최종 성과를 개선한다. 학생. 학생 중간 은닉층은 일반적으로 교사의 중간 은닉층보다 작기 때문에 학생 은닉층을 교사 은닉층의 예측에 매핑하기 위해 추가 매개변수가 도입됩니다. 이를 통해 더 나은 일반화 또는 더 빠르게 실행할 수 있는 더 깊은 학생을 훈련할 수 있으며, 이는 선택한 학생의 능력에 의해 제어됩니다. 예를 들어, CIFAR-10에서 매개변수가 거의 10.4배 적은 심층 학생 네트워크는 더 큰 최신 교사 네트워크보다 성능이 뛰어납니다.
The Foreign Exchange Market is the biggest and one of the most liquid markets in the world. This market has always been one of the most challenging markets as far as short term prediction is concerned. Due to the chaotic, noisy, and non-stationary nature of the data, the majority of the research has been focused on daily, weekly, or even monthly prediction. The literature review revealed that there is a gap for intra-day market prediction. Identifying this gap, this paper introduces a prediction and decision making model based on Artificial Neural Networks (ANN) and Genetic Algorithms. The dataset utilized for this research comprises of 70 weeks of past currency rates of the 3 most traded currency pairs: GBP\USD, EUR\GBP, and EUR\USD. The initial statistical tests confirmed with a significance of more than 95% that the daily FOREX currency rates time series are not randomly distributed. Another important result is that the proposed model achieved 72.5% prediction accuracy. Furthermore, implementing the optimal trading strategy, this model produced 23.3% Annualized Net Return.
In this paper, a novel decision support system using a computational efficient functional link artificial neural network (CEFLANN) and a set of rules is proposed to generate the trading decisions more effectively. Here the problem of stock trading decision prediction is articulated as a classification problem with three class values representing the buy, hold and sell signals. The CEFLANN network used in the decision support system produces a set of continuous trading signals within the range 0 to 1 by analyzing the nonlinear relationship exists between few popular technical indicators. Further the output trading signals are used to track the trend and to produce the trading decision based on that trend using some trading rules. The novelty of the approach is to engender the profitable stock trading decision points through integration of the learning ability of CEFLANN neural network with the technical analysis rules. For assessing the potential use of the proposed method, the model performance is also compared with some other machine learning techniques such as Support Vector Machine (SVM), Naive Bayesian model, K nearest neighbor model (KNN) and Decision Tree (DT) model.
The motivation behind this research is to innovatively combine new methods like wavelet, principal component analysis (PCA), and artificial neural network (ANN) approaches to analyze trade in today’s increasingly difficult and volatile financial futures markets. The main focus of this study is to facilitate forecasting by using an enhanced denoising process on market data, taken as a multivariate signal, in order to deduct the same noise from the open-high-low-close signal of a market. This research offers evidence on the predictive ability and the profitability of abnormal returns of a new hybrid forecasting model using Wavelet-PCA denoising and ANN (named WPCA-NN) on futures contracts of Hong Kong’s Hang Seng futures, Japan’s NIKKEI 225 futures, Singapore’s MSCI futures, South Korea’s KOSPI 200 futures, and Taiwan’s TAIEX futures from 2005 to 2014. Using a host of technical analysis indicators consisting of RSI, MACD, MACD Signal, Stochastic Fast %K, Stochastic Slow %K, Stochastic %D, and Ultimate Oscillator, empirical results show that the annual mean returns of WPCA-NN are more than the threshold buy-and-hold for the validation, test, and evaluation periods; this is inconsistent with the traditional random walk hypothesis, which insists that mechanical rules cannot outperform the threshold buy-and-hold. The findings, however, are consistent with literature that advocates technical analysis.
This paper propose that the combination of smoothing approach taking into account the entropic information provided by Renyi method, has an acceptable performance in term of forecasting errors. The methodology of the proposed scheme is examined through benchmark chaotic time series, such as Mackay Glass, Lorenz, Henon maps, the Lynx and rainfall from Santa Francisca series, with addition of white noise by using neural networks-based energy associated (EAS) predictor filter modified by Renyi entropy of the series. In particular, when the time series is short or long, the underlying dynamical system is nonlinear and temporal dependencies span long time intervals, in which this are also called long memory process. In such cases, the inherent nonlinearity of neural networks models and a higher robustness to noise seem to partially explain their better prediction performance when entropic information is extracted from the series. Then, to demonstrate that permutation entropy is computationally efficient, robust to outliers, and effective to measure complexity of time series, computational results are evaluated against several non-linear ANN predictors proposed before to show the predictability of noisy rainfall and chaotic time series reported in the literature.
This study presents a neural network & web-based decision support system (DSS) for foreign exchange (forex) forecasting and trading decision, which is adaptable to the needs of financial organizations and individual investors. In this study, we integrate the back-propagation neural network (BPNN)- based forex rolling forecasting system to accurately predict the change in direction of daily exchange rates, and the Web-based forex trading decision support system to obtain forecasting data and provide some investment decision suggestions for financial practitioners. This research reveals the structure of the DSS by the description of an integrated framework, and meantime we find that the DSS is integrated, user-oriented by its implementation, and practical applications reveal that this DSS demonstrates very high forecasting accuracy and its trading recommendations are reliable.
The Foreign Exchange Market is the biggest and one of the most liquid markets in the world. This market has always been one of the most challenging markets as far as short term prediction is concerned. Due to the chaotic, noisy, and non-stationary nature of the data, the majority of the research has been focused on daily, weekly, or even monthly prediction. The literature review revealed that there is a gap for intra-day market prediction. Identifying this gap, this paper introduces a prediction and decision making model based on Artificial Neural Networks (ANN) and Genetic Algorithms. The dataset utilized for this research comprises of 70 weeks of past currency rates of the 3 most traded currency pairs: GBP\USD, EUR\GBP, and EUR\USD. The initial statistical tests confirmed with a significance of more than 95% that the daily FOREX currency rates time series are not randomly distributed. Another important result is that the proposed model achieved 72.5% prediction accuracy. Furthermore, implementing the optimal trading strategy, this model produced 23.3% Annualized Net Return.