Scribd
Upload
30 views3 pages

MScFE 690 CP Problem Statement Guidelines

The document provides guidelines for the MScFE 690 Capstone project, emphasizing the importance of clearly scoping the problem statement for successful completion. It includes examples of specific research topics related to machine learning models for cryptocurrency forecasting and outlines the assessment criteria for the chosen models. Additionally, it details the grading rubric for the problem statement, goals, objectives, and writing quality.

Uploaded by

Shalabh Tewari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
30 views3 pages

MScFE 690 CP Problem Statement Guidelines

The document provides guidelines for the MScFE 690 Capstone project, emphasizing the importance of clearly scoping the problem statement for successful completion. It includes examples of specific research topics related to machine learning models for cryptocurrency forecasting and outlines the assessment criteria for the chosen models. Additionally, it details the grading rubric for the problem statement, goals, objectives, and writing quality.

Uploaded by

Shalabh Tewari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

MScFE 690 CAPSTONE

Problem Statement Guidelines (M2)


See grading rubric here.

Academic research is one of the most fulfilling experiences. At WQU, the purpose of the
final capstone project is to encourage you to use your interest, creativity, and knowledge
to delve as deep into the topic as you want.

Researching any topic can become a complex maze if students do not properly scope
their problem of interest. To help you successfully complete assignments M1 and M2
for your capstone project, the following guidelines have been created.

Note that M2 is particularly crucial because this is where you map out the exact
destination of your capstone project. Thus, we encourage you to think thoroughly about
these two assignments as that will make the rest of your project quite easy.

For the example below we are assuming that a student is interested in the following
topic: “ Applying Machine Learning models to forecast cryptocurrencies”. Analogy can
be drawn to any topic of your choice. Please connect with your instructor within the
first week (if not before) to see how you should scope the problem.

Example
Applying machine learning models to forecast asset prices is a fairly exciting feat but the
number of models (machine learning or otherwise) are quite a few and you should know
exactly which model you would like to try. Saying Machine learning is NOT Specific; nor
is deep learning.

Using the example above, let’s narrow down the hypothesis to one of the following
(these are just examples; you can pick another topic)

1. Apply Linear regression model to daily time series to predict the prices of
BTCUSDT using daily data points.

MScFE | © 2023 - WorldQuant University – All rights reserved. Group Work Project 2 | PAGE 1
2. Compare Neural Networks to Support Vector Machines to see which model
predicts the price of ETHUSDT on a 4 hour candlestick chart.
3. Apply LSTM Deep learning model to 1 minute ETHBTC prices to predict the
confidence interval of prices over a 10 minute horizon.

For each of the specified thesis above, you will find the following commanilites.

1. They each specify a model they want to use to forecast. Or they specify the two
specific models they want to compare. But the models are specified.
2. They specify the asset price they want to forecast. Rather than saying
Cryptocurrency, a specific symbol is specified.
3. A time horizon is specified. Whether its daily data, or 1 minute, there is a clear
timeseries of the data to be analyzed.

Model Assessment

Once you have chosen the model, please research how the model will be assessed. For
instance, is Root Mean Squared Error good enough to test the model? If you are
devising a strategy based on your model then you should consider dividing the data set
into 2; Backtesting and walk forward analysis. You may consider a library like Vector BT.
For a trading strategy all the following are expected outputs to assess its effectiveness.

MScFE | © 2023 - WorldQuant University – All rights reserved. Group Work Project 2 | PAGE 2
Problem Statement - Grading Rubric
(7.4% or 50 + 24 points for group collaboration)

Problem Statement Goals and Objectives Writing and Formatting

30 Points 15 Points 5 Points

The Problem Statement is clear The Problem Statement lists A submission that looks
and provides context and comprehensive goals and professional should include:
consequences that indicate the objectives for the project being
importance and relevance of the developed. ● Axes, labels, and scales in
project. graphs
● No significant grammar
The Problem Statement also errors or typos
includes the initial code design or ● Organized, clear structure,
pseudo code, or at least a code and easy to read document
design plan that outlines how ● Proper citations and
your Group intends to develop the bibliography using MLA
code. format

Revision date: December 10, 2024

MScFE | © 2023 - WorldQuant University – All rights reserved. Group Work Project 2 | PAGE 3

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy