Матлаб И Форекс

In this webinar we will present an example workflow for researching, implementing and testing an automated trading strategy. You will learn how MATLAB^® and add-on products can be used for data gathering, preparation and visualization, model development, backtesting, calibration, integration with existing systems and ultimately deployment.
We look at each of the parts in this process and see how MATLAB provides a single platform that allows the efficient solution of all parts of this problem.

As of Ra, Trading Toolbox has been merged into Datafeed Toolbox. As a part of this merge, a subset of the functionality is moving to File Exchange. Find more details in the release notes.

Specific topics include:

• Data gathering options, including daily historic, intraday, and real-time data
• Model building and prototyping in MATLAB
• Backtesting and calibrating a model
• Interacting with existing libraries and software for trade execution
• Deployment of the final application in a number of environments, including .NET, JAVA, and Excel
• Tools for high frequency trading, including parallel computing, GPUs, and C code generation from MATLAB

View example code from this webinar.

About the Presenter: Stuart Kozola is a product manager at MathWorks and focuses on MATLAB^® and add-on products for computational finance. Prior to joining MathWorks in , Stuart worked at Pratt & Whitney (United Technologies) as a design engineer working on combustion systems for gas turbine engines. Stuart earned a B.S. in Chemical Engineering from the University of Wyoming, M.S. in Chemical Engineering from Arizona State University, M.S. in Electrical Engineering from Rensselaer Polytechnic Institute, and an M.B.A. from Carnegie Mellon University.

Overview

In this webinar we will use regression and machine learning techniques in MATLAB to train and test an algorithmic trading strategy on a liquid currency pair. Using real life data, we will explore how to manage time-stamped data, create a series of derived features, then build predictive models for short term FX returns.

We will then show how to backtest this strategy historically, while taking into account trading costs in the strategy and the machine learning modelling process.

Highlights

• Handling data using the timetable object
• Linear regression modelling
• Machine Learning techniques for Supervised Learning
• Backtesting strategy performance historically

About the Presenter

Dan Owen is Industry Manager for Financial Applications for the APAC region. Dan has worked at MathWorks for over 12 years in Consulting and as an Applications Engineer, always focusing on Financial Services. He has also worked as a Director of Systematic Trading at Dresdner Kleinwort and within a Quant Technology group at Fidelity International. He holds a BSc and a PhD in Applied Mathematics from the University of Birmingham in the United Kingdom.

Machine Learning with Matlab Final project on Timeseries Prediction with LSTM / RNN.

1. Project directory structure:

ml_proj/ +-- data % Contains EURUSD dataset +-- scripts % MATLAB Scripts +-- measure % RMSE and MAPE implementation

getting online Forex data

Sie verfolgen jetzt diese Frage

17 Ansichten (letzte 30 Tage)

Good evening everyone,

i have started a new work that i need to get an online rate of Euro to USD.

and i just use this code for this purpose,

apikey = '***********';

c = quandl(apikey);

s = 'WGC/GOLD_DAILY_USD';

startdate = datetime('','InputFormat','MM-dd-yyyy');

enddate = datetime('now','InputFormat','MM-dd-yyyy');

periodicity = 'daily';

d = history(c,s,startdate,enddate,periodicity,

"limit",inf,"order","asc");

but the problem is this code can only give me daily and monthly data but i need tick, minutes and hourly data too.

so the question is, is there any way to fetch an online rate of Euro to USD (from anywhere freely or not) in the Matlab environment?

thanks in advance,

Abolfazl

Antworten (1)

Bearbeitet: Saeid Kamandiam 29 Dez.

Hi, Abolfazl, to get forex data, you need to use MetaTrader 4, the expert data export v3 can also help you, but this is the manual way to get the data.

Es ist ein Fehler aufgetreten

Da Änderungen an der Seite vorgenommen wurden, kann diese Aktion nicht abgeschlossen werden. Laden Sie die Seite neu, um sie im aktualisierten Zustand anzuzeigen.

Translated by

Website auswählen

Wählen Sie eine Website aus, um übersetzte Inhalte (sofern verfügbar) sowie lokale Veranstaltungen und Angebote anzuzeigen. Auf der Grundlage Ihres Standorts empfehlen wir Ihnen die folgende Auswahl: .

Sie können auch eine Website aus der folgenden Liste auswählen:

So erhalten Sie die bestmögliche Leistung auf der Website

Wählen Sie für die bestmögliche Website-Leistung die Website für China (auf Chinesisch oder Englisch). Andere landesspezifische Websites von MathWorks sind für Besuche von Ihrem Standort aus nicht optimiert.

Kontakt zu Ihrer lokalen Niederlassung

1. Installation In order to run the project, the following Matlab Toolboxes must be installed: 1. Statistics and Machine Learning Toolbox 2. Econometrics Toolbox 3. Deep Learning Toolbox

2. Running the Project

   • Add ml_proj and its subfolders into path
   • Modify the configuration if necessary. By default, it only verifies the saved models in full dataset. Set eunic-brussels.eude to "train" to train the model again.
   • Run the respective .m file