G. Peter Zhang Neurocomputing 50 (2003) 159–175 link Time series forecasting using a hybrid ARIMA and neural network model Presented by Trent Goughnour.

Slides:



Advertisements
Présentations similaires
Modèles et observations Confidence in the validity of a finding is based on the type, amount, quality, and consistency of evidence (e.g., data, mechanistic.
Advertisements

FINANCE Distribution des rentabilités Professeurr André Farber Solvay Business School Université Libre de Bruxelles.
The subjunctive mood If I were you, I’d call him It is absolutely necessary that you be there on time May God save the queen! Normally, in English we would.
Making PowerPoint Slides Avoiding the Pitfalls of Bad Slides.
PERFORMANCE One important issue in networking is the performance of the network—how good is it? We discuss quality of service, an overall measurement.
An Introduction To Two – Port Networks The University of Tennessee Electrical and Computer Engineering Knoxville, TN wlg.
Traffic Sign Recognition Jacob Carlson Sean St. Onge Advisor: Dr. Thomas L. Stewart.
 Components have ratings  Ratings can be Voltage, Current or Power (Volts, Amps or Watts  If a Current of Power rating is exceeded the component overheats.
IP Multicast Text available on
Template Provided By Genigraphics – Replace This Text With Your Title John Smith, MD 1 ; Jane Doe, PhD 2 ; Frederick Smith, MD, PhD 1,2 1.
Overview of SUN’s Unix Campus-Booster ID : **XXXXX Copyright © SUPINFO. All rights reserved Introduction to Solaris 10.
Look at these two pictures.  Do you recognize these two places?  Would you be able to tell me where each picture was taken?  Why?
Globalization
Exploration on statistical and time-series models
Theme Three Speaking Questions
L'heure.
Speaking Exam Preparation
La Santé Unité 10.
CONNAÎTRE ET SAVOIR les notes.
LEACH & LEACH-C & V-LEACH. Overview Introduction Existing Protocols – Direct Transmission – Minimum Transmission Energy LEACH Stochastic Threshold Algorithm.
Qu’est-ce qu’ils aiment faire?
Mon enfance Quand tu étais petit, tu étais comment?
1 S Transmission Methods in Telecommunication Systems (4 cr) Transmission Channels.
Technologies de l’intelligence d’affaires Séance 14
Le futur antérieur Future Perfect.
Strengths and weaknesses of digital filtering Example of ATLAS LAr calorimeter C. de La Taille 11 dec 2009.
Quantum Computer A New Era of Future Computing Ahmed WAFDI ??????
Les questions et les mots interrogatifs
Point de départ Relative pronouns link two phrases together into a longer, more complex sentence. The second phrase gives additional information about.
Statistics & Econometrics Statistics & Econometrics Statistics & Econometrics Statistics & Econometrics Statistics & Econometrics Statistics & Econometrics.
Theme One Speaking Questions
Arch. Hydrobiol. I 100 I I Stuttgart, Mai 1984
Notes: l’Imparfait français III.
About INTEGRA The Integrated community, probation and prison services radicalisation prevention approach strives to improve the transition process between.
POPULATION GROWTH IN AFRICA EXHIBITOR: Papa Abdoulaye Diouf.
1. Financial (Accounting) Statements  Financial or Accounting statements are used for reporting corporate activity. 2 For Stakeholders.
P&ID SYMBOLS. P&IDs Piping and Instrumentation Diagrams or simply P&IDs are the “schematics” used in the field of instrumentation and control (Automation)
Leadership Styles Mrs. Keith Main Types of Leadership Styles 1.The Autocratic or Authoritarian Leader 2.The Democratic or Participative Leader.
Author : Moustapha ALADJI PhD student in economics-University of Guyana Co-author : Paul ROSELE Chim HDR Paris 1-Pantheon Sorbonne Economics / Management.
Essai
Distributed Radiation Detection Daniel Obenshain Arthur Rock SURF Fellow.
Introduction to Computational Journalism: Thinking Computationally JOUR479V/779V – Computational Journalism University of Maryland, College Park Nick Diakopoulos,
High-Availability Linux Services And Newtork Administration Bourbita Mahdi 2016.
Benchmarking noise policies
Information available in a capture history
Localization problems for an AUV
CS 790Q Biometrics Face Recognition Using Dimensionality Reduction PCA and LDA M. Turk, A. Pentland, "Eigenfaces for Recognition", Journal of Cognitive.
Definition Division of labour (or specialisation) takes place when a worker specialises in producing a good or a part of a good.
Roots of a Polynomial: Root of a polynomial is the value of the independent variable at which the polynomial intersects the horizontal axis (the function.
Quelle est la date aujourd’hui?
Understanding Constraints in Primavera P6Primavera P6.
1-1 Introduction to ArcGIS Introductions Who are you? Any GIS background? What do you want to get out of the class?
By:- Israr K. Raja Islamabad, Pakistan. Supply Chain Activities those Affect the Financial Performance Supply chain managers make decisions and use organizational.
Les Verbes Réfléchis What you do to yourself.
What’s the weather like?
5 Whys A technique for getting past first impressions and finding the root cause of a problem.
USE OF PASSIVE PASSIVE VOICE IS USED WHEN THE FOCUS IS ON THE ACTION. IT IS NOT IMPORTANT OR NOT KNOWN, HOWEVER, WHO OR WHAT IS PERFORMING THE ACTION.
Making PowerPoint Slides Avoiding the Pitfalls of Bad Slides.
POWERPOINT PRESENTATION FOR INTRODUCTION TO THE USE OF SPSS SOFTWARE FOR STATISTICAL ANALISYS BY AMINOU Faozyath UIL/PG2018/1866 JANUARY 2019.
By : HOUSNA hebbaz Computer NetWork. Plane What is Computer Network? Type of Network Protocols Topology.
5S Methodology How to implement "5S" and get extraordinary results.
1 Sensitivity Analysis Introduction to Sensitivity Analysis Introduction to Sensitivity Analysis Graphical Sensitivity Analysis Graphical Sensitivity Analysis.
Si clauses in French.
Avoiding the Pitfalls of Bad Slides Tips to be Covered Outlines Slide Structure Fonts Colour Background Graphs Spelling and Grammar Conclusions Questions.
Le Passé Composé (Perfect Tense)
Les opinions Les opinions = Opinions. In this lesson pupils will learn to understand and give their own opinions about singular items.
Soutenance de thèse: Okba Taouali 1 02/08/2019 Fathia AZZOUZI, Adam BOURAS, Nizar JEBLI Conceptual specifications of a cooperative inter- machines dialogue.
Over Sampling methods IMBLEARN Package Realised by : Rida benbouziane.
IMPROVING PF’s M&E APPROACH AND LEARNING STRATEGY Sylvain N’CHO M&E Manager IPA-Cote d’Ivoire.
Transcription de la présentation:

G. Peter Zhang Neurocomputing 50 (2003) 159–175 link Time series forecasting using a hybrid ARIMA and neural network model Presented by Trent Goughnour Illinois State Department of Mathematics

Background Methodology Data Results Conclusion Overview

Forecasting Past observations to develop a model Model is then used to forecast future values Linear Methods  Auto Regressive  Moving Average  Exponential smoothing Non-Linear Methods  Bilinear model  Threshold autoregressive (TAR) model  Autoregressive conditional heteroskedastic (ARCH)  More recently artificial neural networks (ANN) and other machine learning Traditional Time series forecasting models

Autoregressive Integrated Moving Average (ARIMA) Models: Refer to models where the dependent variable depends on its own past history as well as the past history of random shocks to its process. Auto Regressive (AR) Integrated (I) Moving Average (MA) An ARIMA(p, d, q) is represented by three parameters: p, d, and q, where p is the degree of autoregressive, d is the degree of integration, and q is the degree of moving average. ARIMA

ARIMA Examples

Artificial Neural Networks

Hybrid Approach

ARIMA is implemented in this paper using SAS/ETS systems ANN models are built using Generalize Reduced Gradient Algorithm (GRG2). GRG2 based training system is used for this portion. Side note that both of these are available in R. Implementation

Three well-known data sets  the Wolf’s sunspot data  the Canadian lynx data  the British pound/US dollar exchange rate Data Sample compositions in three data sets SeriesSample sizeTraining set (size)Test set (size) Sunspot –1920 (221) (253) 1921–1987 (67) (35) Lynx –1920 (100)1921–1934 (14) Exchange rate –1992 (679)1993 (52)

Data Visualized Weekly BP=USD exchange rate series (1980–1993) Canadian lynx series ( ) Sunspot series (1700–1987)

ModelMSEMAD 35 aheadARIMA ANN Hybrid aheadARIMA ANN Hybrid Sunspot Results 35-period forecasts for hybrid are 16.13% better MSE than ARIMA 67-period not as good, but still better predictions.

Sunspot Results

ModelMSEMAD ARIMA ANN Hybrid Lynx Results 18.87% decrease in MSE 7.97% improvement in MAD

Lynx Results

ModelMSEMAD 1 monthARIMA ANN Hybrid monthARIMA ANN Hybrid monthARIMA ANN Hybrid Pound/Dollar Conversion Shows improvement across three different time horizons. ARIMA model shows that a simple random walk is the best model

Tuning of neural network was done to get optimal predictions 4x4x1 network for sunspot data 7x5x1 for lynx data 7x6x1 for exchange rate data ARIMA for exchange rate becomes random walk Additional Results

Artificial neural nets alone seem to be an improvement over standard ARIMA. The empirical results with three real data sets clearly suggest that the hybrid model is able to outperform each component model used in isolation. Conclusions

Theoretical as well empirical evidences suggests using dissimilar models or models that disagree with each other strongly, the hybrid model will have lower generalization variance or error. using the hybrid method can reduce the model uncertainty fitting the ARIMA model first to the data, the overfitting problem that is related to neural network models can be eased. Conclusions cont.