scholarly journals Qual Var Revisited: Good Forecast, Bad Story

2016 ◽  
Vol 19 (2) ◽  
pp. 293-321 ◽  
Author(s):  
Makram El-Shagi ◽  
Gregor Von Schweinitz
Keyword(s):  
Good Forecast

Parameter C Optimizing for Robust ε-Support Vector Regression

2014 ◽  
Vol 543-547 ◽  
pp. 2045-2048
Author(s):  
Yuan Lv ◽  
Zhong Gan
Keyword(s):  
Experimental Data ◽  
Support Vector Regression ◽  
System Parameter ◽  
Forecast Accuracy ◽  
Regression Method ◽  
Support Vector ◽  
Penalty Parameter ◽  
Optimal Method ◽  
Good Forecast ◽  
Selection Of

In case of experimental data contaminated with errors and noise, the robust ε-support vector regression has good forecast accuracy and high generalization ability. However, it depends on the selection of system parameter. Firstly, this paper introduces the robust ε-support vector regression method. Secondly, as the experiments prove, the new method achieves high forecast accuracy by virtue of the optimal penalty parameter C. Finally, the optimal method of parameter C is presented in the last section.

A Method of Improved Support Vector Machine for Network Security Situation Forecasting

2011 ◽  
Vol 187 ◽  
pp. 291-296
Author(s):  
Yuan Cheng Li ◽  
Jing Tao Jing
Keyword(s):  
Network Security ◽  
Forecast Accuracy ◽  
Forecast Model ◽  
Selection Method ◽  
Support Vector ◽  
Swarm Optimization ◽  
Parameters Selection ◽  
Vector Machines ◽  
Good Forecast ◽  
Improved Support Vector Machine

Aiming at the problem that parameters of Support Vector Machines (SVM) are very difficult to confirm, this paper points out a parameter selection method for SVM based on Particle Swarm Optimization (PSO), which can make the SVM more scientific and reasonable in parameters selection; and thus enhance the forecast accuracy of the network security situation. The Simulation results show that the optimized SVR forecast model has good forecast accuracy for the network security situation, and present the future changing at a macro level, then help the network managers control network.

scholarly journals Optimasi Parameter Pemulusan Algoritma Brown Menggunakan Metode Golden Section Untuk Prediksi Data Tren Positif dan Negatif

2018 ◽  
Vol 2 (1) ◽  
pp. 307-314 ◽  
Author(s):  
Fiqih Akbari ◽  
Arief Setyanto ◽  
Ferry Wahyu Wibowo
Keyword(s):  
Golden Section ◽  
Exponential Smoothing ◽  
Double Exponential ◽  
Double Exponential Smoothing ◽  
Good Forecast

Algoritma DES (Double Exponential Smoothing) Brown merupakan algoritma peramalan yang digunakan untuk memprediksi data deret berkala baik berpola tren positif maupun tren negatif. Namun algoritma ini mempunyai kelemahan yaitu dalam menentukan nilai parameter optimum untuk meminimasi error peramalan (MAPE), nilai parameter tersebut dicari menggunakan metode Golden Section dimana sebelumnya dicari secara manual menggunakan percobaan berulang kali. Penelitian ini menggunakan 60 data berpola tren yang dianalisis untuk pengelompokan pola data tren positif dan negatif dimana selanjutnya dilakukan proses peramalan, evaluasi dan pengujian untuk mengetahui jenis pola data tren apa yang terbaik. Dari hasil perhitungan dan pengujian diketahui bahwa parameter optimasi menghasilkan nilai MAPE yang optimum, dimana selanjutnya nilai parameter tersebut dilakukan proses peramalan pada kelompok pola data tren positif dan negatif yang menghasilkan rata-rata nilai MAPE sebesar 9,73401% (highly accurate) untuk data berpola tren positif dan 15,78467% (good forecast) untuk data berpola tren negatif. Algoritma peramalan DES Brown dengan metode optimasi parameter menghasilkan nilai pendekatan terhadap data asli jika data tersebut menunjukkan penambahan atau penurunan nilai disekitar nilai rata-rata. Sebaliknya, akan menghasilkan nilai MAPE yang tinggi (tidak akurat) jika data tersebut memiliki lonjakan periode nilai data. Dari kedua kelompok nilai MAPE tersebut dilakukan uji t statistik yang menyatakan bahwa data berpola tren positif () menghasilkan nilai rata-rata MAPE lebih baik dibandingkan data berpola tren negatif (μ2).  

scholarly journals Multistep Flood Inundation Forecasts with Resilient Backpropagation Neural Networks: Kulmbach Case Study

Water ◽  
2020 ◽  
Vol 12 (12) ◽  
pp. 3568
Author(s):  
Qing Lin ◽  
Jorge Leandro ◽  
Stefan Gerber ◽  
Markus Disse
Keyword(s):  
Neural Networks ◽  
Flood Hazard ◽  
Forecast Accuracy ◽  
Threshold Value ◽  
Flood Inundation ◽  
Flood Events ◽  
Forecast Performance ◽  
Flooding Event ◽  
Good Forecast

Flooding, a significant natural disaster, attracts worldwide attention because of its high impact on communities and individuals and increasing trend due to climate change. A flood forecast system can minimize the impacts by predicting the flood hazard before it occurs. Artificial neural networks (ANN) could efficiently process large amounts of data and find relations that enable faster flood predictions. The aim of this study is to perform multistep forecasts for 1–5 h after the flooding event has been triggered by a forecast threshold value. In this work, an ANN developed for the real-time forecast of flood inundation with a high spatial resolution (4 m × 4 m) is extended to allow for multiple forecasts. After trained with 120 synthetic flood events, the ANN was first tested with 60 synthetic events for verifying the forecast performance for 3 h, 6 h, 9 h and 12 h lead time. The model produces good results, as shown by more than 81% of all grids having an RMSE below 0.3 m. The ANN is then applied to the three historical flood events to test the multistep inundation forecast. For the historical flood events, the results show that the ANN outputs have a good forecast accuracy of the water depths for (at least) the 3 h forecast with over 70% accuracy (RMSE within 0.3 m), and a moderate accuracy for the subsequent forecasts with (at least) 60% accuracy.

Using certain indices of fibrinolytic system for forecasting the efficiency of neoajuvant chemotherapy in ovarian cancer.

2013 ◽  
Vol 31 (15_suppl) ◽  
pp. e22022-e22022
Author(s):  
Yelena Frantsiyants ◽  
Tatiana Moiseyenko ◽  
Yekaterina Komarova ◽  
Larisa Kozlova
Keyword(s):  
Ovarian Cancer ◽  
Clinical Effect ◽  
Tumor Tissue ◽  
Biological Properties ◽  
Fibrinolytic System ◽  
Individual Sensitivity ◽  
Specific Therapy ◽  
Efficient Treatment ◽  
Further Development ◽  
Good Forecast

e22022 Background: Using cell markers that characterize biological properties of a tumor, the relevance of treatment used can be determined for each patient, further ways of surveillance of the patients can be selected, and individual forecasts of the illness can be made. Methods: After 1 course of neoajuvant chemoterapy according to the scheme CAP, in the malign tumor tissue of 21 patients (T3a-c N0-1M0-1), the activity of plasminogen and plasmin was determined by ELISA. Then the ratio K of plasminogen (PG) and plasmin (P) was calculated. As a blank, tumor tissue of 18 patients having ovarian cancer without chemotherapy administered has been studied. On all patients, operations in full or cytoreductive surgery were performed. Results: With efficient treatment, ratio K increased (К=1,4±0,2) relative to indices in patients without neoajuvant chemotherapy (К=0,3±0,17), while it remained the same (К=0,2±0,09) or decreased relative to initial data if there was no clinical effect, which is confirmed by further development of ovarian cancer. Conclusions: Thus, determining of plasminogen and plasmin – fibrinolytic system indices – in the tumor tissue and calculating of their ratio allows detecting the patients who have effect from treatment and good forecast and ones without effect, having bad forecast, who need aggressive ajuvant treatment. It also allows determining individual sensitivity or resistance of malign tumors in patients to certain regimens of chemotherapy. At the same time, there is an opportunity to objectify the results of specific therapy.

Technology Forecasting Methods with Emphasis on Bibliographic Analysis and Curve Fitting: A Photocatalytic Case Example

2012 ◽  
Vol 442 ◽  
pp. 144-148 ◽  
Author(s):  
Promsak Jaranyagorn ◽  
Chuvej Chansa Ngavej
Keyword(s):  
Decision Making ◽  
Curve Fitting ◽  
New Products ◽  
Technology Change ◽  
Technology Forecasting ◽  
Bibliographic Analysis ◽  
Strategic Direction ◽  
Forecasting Methods ◽  
Trend Forecasting ◽  
Good Forecast

Technology forecasting methods can be applied to make sure we know the potential direction, characteristic, state and effect of technology change. A good forecast can maximize gain and minimize loss from future conditions Nowadays, many companies invest a great deal in R&D to develop competitive new products and technology. Globalization and the rate of technological change in highly competitive market mean that companies need to consider increasing the R&D budgets and ensuring the money is spent efficiently and effectively. Technology forecasting is very useful for decision making in managerial issues. It can help government manage their public agendas and budgetary constraints and for business strategic direction and prioritizing R&D projects. This paper explores the technology forecasting methods and demonstrates the use of bibliographic analysis and curve fitting with Bass diffusion and exponential models for trend forecasting of titanium dioxide photocatalyst as a case example.

scholarly journals Orographic cyclogenesis in a saturated atmosphere and intense precipitation: baroclinic modal solutions under the joint action of localized mountains and humidity

1997 ◽  
Vol 40 (6) ◽  
Author(s):  
R. Benzi ◽  
M. Fantini ◽  
R. Mantovani ◽  
A. Speranza
Keyword(s):  
Unstable Mode ◽  
Space Structure ◽  
Simplified Model ◽  
Modal Theory ◽  
Intense Precipitation ◽  
Saturated Atmosphere ◽  
Atmospheric Motion ◽  
Baroclinic Modes ◽  
Good Forecast ◽  
Precipitation Events

In this paper we analyse the nature of orographic cyclogenesis in a saturated atmosphere by means of a simplified model based on the analysis of linear modal solutions.The space structure of fastest growing modal solutions suggests that three different scales of axtratropical atmospheric motion may simultaneously be activated in a single, growing, unstable mode: the orographic modulation of growing baroclinic modes extending, as we know from the classical modal theory of orographic cyclogenesis, from the scale typical of the primary, extra-tropical cyclone to the scale of the secondary, orographic cyclone, is also characterized by the (smaller) scale associated with strong ascending motion in a saturated atmosphere. Since ascending motion can be associated with intense precipitation, this result is important in view of its potential consequences both on the ability to achieve a good forecast of intense precipitation events in the Mediterranean and on the refinement of the theory of orographic cyclogenesis.

Robustness Analysis of Novel ε-Support Vector Regression

2014 ◽  
Vol 543-547 ◽  
pp. 2049-2052
Author(s):  
Yuan Lv ◽  
Zhong Gan
Keyword(s):  
Normal Distribution ◽  
Support Vector Regression ◽  
Robustness Analysis ◽  
Forecast Accuracy ◽  
Support Vector ◽  
Kolmogorov Test ◽  
Hyper Plane ◽  
Good Forecast

The key to the robust ε-support vector regression algorithm is searching for the optimal regression hyper plane while data with disturbance in the X-direction. In the paper, the optimal regression hyper plane and the optimal separating hyper plane are compared and analyzed. By means of Kolmogorov test, it is can be deduced that the testing errors of the robust ε-support vector regression experiments follow normal distribution. The result demonstrates that the algorithm has good forecast accuracy and high robustness.

Application of Projection Pursuit Regression in Equipment Indemnificatory Valuation and Forecast

2013 ◽  
Vol 411-414 ◽  
pp. 2111-2114
Author(s):  
Lian Jun Zhu ◽  
Hong Yan Li ◽  
Yu Cai Dong ◽  
Tian Yuan Jiang ◽  
Ge Hua Fan
Keyword(s):  
Regression Model ◽  
Relative Error ◽  
Projection Pursuit ◽  
Influential Factors ◽  
Average Relative Error ◽  
Projection Pursuit Regression ◽  
Good Balance ◽  
Training Samples ◽  
Good Forecast

The Theory of Projection Pursuit Regression is applied in the equipment indemnificatory valuation and forecast to establish the projection pursuit regression model. After fitting the training samples, this model strikes a good balance between the valuation value and its relevant influential factors, demonstrating a good fitting effect with the average relative error of only 2.1522% . After predicting the test samples, it shows a good forecast effect with the relative error of only-0.4069%, thus providing basis for equipment indemnificatory valuation and forecast.

scholarly journals Application of SARIMA Model for Forecasting Consumption of Electricity in Gezira State, Sudan (2006-2018): تطبيقات نماذج ساريما للتنبؤ باستهلاك الكهرباء في ولاية الجزيرة, السودان  (2006- 2018)

2019 ◽  
Vol 3 (4) ◽  
Author(s):  
Nada Mohammed Ahmed Alamin
Keyword(s):  
Historical Data ◽  
Moving Average ◽  
Electricity Consumption ◽  
Seasonal Behavior ◽  
Sarima Model ◽  
Control Center ◽  
Autoregressive Integrated Moving Average ◽  
Good Forecast ◽  
National Control ◽  
Monthly Electricity Consumption

    The purpose of the research is to reach the forecast of monthly electricity consumption in Gezira state, Sudan for the period (Jun 2018 - Dec 2020) through the application to the historical data of electric power consumption (Jan 2006-May 2018) obtained from the National Control Center, which has been applied in the research methodology of seasonal Autoregressive Integrated Moving Average due to seasonal behavior in the data, good forecast has been given by SARIMA (2, 1, 7) (0, 1, 1), which has been examined its quality using the Thiel coefficient. The study recommended the use of the model of seasonal Autoregressive Integrated Moving Average in data with Seasonal behavior due to its simple application and accuracy of the results reached.    

Sign in / Sign up

Export Citation Format

Share Document