Next Article in Journal
The Impact of Tax Preferences on the Investment Attractiveness of Bonds for Retail Investors: The Case of Russia
Next Article in Special Issue
Impact Analysis of Financial Regulation on Multi-Asset Markets Using Artificial Market Simulations
Previous Article in Journal
Dynamic Spillovers between Gulf Cooperation Council’s Stocks, VIX, Oil and Gold Volatility Indices
Previous Article in Special Issue
The Predictability of the Exchange Rate When Combining Machine Learning and Fundamental Models
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JRFM
Volume 13
Issue 4
10.3390/jrfm13040071
Review Report
jrfm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

Printed Edition

A printed edition of this Special Issue is available at MDPI Books....
share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Deep Reinforcement Learning in Agent Based Financial Market Simulation

J. Risk Financial Manag. 2020, 13(4), 71; https://doi.org/10.3390/jrfm13040071
by Iwao Maeda 1,*, David deGraw 2, Michiharu Kitano 3, Hiroyasu Matsushima 1, Hiroki Sakaji 1, Kiyoshi Izumi 1 and Atsuo Kato 3
Reviewer 1: Anonymous
Reviewer 2: Anonymous
J. Risk Financial Manag. 2020, 13(4), 71; https://doi.org/10.3390/jrfm13040071
Submission received: 28 February 2020 / Revised: 7 April 2020 / Accepted: 8 April 2020 / Published: 11 April 2020
(This article belongs to the Special Issue AI and Financial Markets)

Round 1

Reviewer 1 Report

This is a very technical paper. Some suggestions on the application of the study should be provided at the conclusion. Here are some suggestions for consideration.

I consider the findings that has great value on Robo-dvisor. Therefore, the authors may consider to mention:

'Robo-advisor serves as financial adviser that provides automated financial advice or investment management for clients. Also, robo-advisor can provide personalized suggestions to clients in more effective ways, while the suggestions can also be updated according to real-time data'

Following two references could be used for the purpose.

Sironi, (2016) FinTech Innovation, 1st ed. Chichester, West Sussex, United Kingdom: Wiley, 2016

Leong, K. and Sung, A., (2018). FinTech (Financial Technology): what is it and how to use technologies to create business value in fintech way?. International Journal of Innovation, Management and Technology, 9(2), pp.74-78.


Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

  1. Remove references [1] and [2] which are not relevant and cite the following studies in Introduction:
  • Intelligent forecasting with machine learning trading systems in chaotic intraday Bitcoin market. Chaos, Solitons & Fractals 133, 109641, 2020.
  • Cryptocurrency forecasting with deep learning chaotic neural networks, Chaos, Solitons & Fractals 118, 35-40, 2020.
  • Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies. Expert Systems with Applications Volume 8315 October 2017 Pages 187-205.
  • Deep learning-based feature engineering for stock price movement prediction. Knowledge-Based Systems Volume 16415 January 2019 Pages 163-173.

 

  1. Why using log-return knowing that artificial neural networks are capable to model both linear and nonlinear data? Argue.
  2. Are there are form of reward functions? Justify why using yours?
  3. Why using two different deep learning models to extract features (LSTM and CNN)? Why not using only CNN which more suitable to extract features not LSTM?
  4. What lessons can we learn from these simulations?
  5. What are the effects on prediction and profit generation on real data? Discuss.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Back to TopTop