[et_pb_section fb_built=”1″ _builder_version=”4.4.5″][et_pb_row _builder_version=”4.4.5″][et_pb_column type=”4_4″ _builder_version=”4.4.5″][et_pb_text _builder_version=”4.9.4″ header_font=”|700||on|||||”]<\/p>\n
[\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row _builder_version=”4.4.5″][et_pb_column type=”4_4″ _builder_version=”4.4.5″][et_pb_text _builder_version=”4.9.4″ header_2_font=”|||on|on||||”]<\/p>\n
This assignment is subject to change up until 3 weeks prior to the due date. We do not anticipate changes; any changes will be logged in this section.<\/span><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row _builder_version=”4.4.5″][et_pb_column type=”4_4″ _builder_version=”4.4.5″][et_pb_text _builder_version=”4.9.4″ header_2_font=”|||on|on||||”]<\/p>\n In this project, you will write software that<\/span> evaluates and prepares portfolio metrics. <\/span>You will submit the code for the project <\/span>in <\/span>Gradescope<\/span> SUBMISSION. You will also submit to Canvas a <\/span>chart as a 1-page report that <\/span>compares<\/span> <\/span>two normalized portfolios<\/span>.<\/span><\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row _builder_version=”4.4.5″][et_pb_column type=”4_4″ _builder_version=”4.4.5″][et_pb_text _builder_version=”4.9.4″ header_2_font=”|||on|on||||”]<\/p>\n Revise the optimization.py code <\/span>to return<\/span> several portfolio statistics: stock allocations (<\/span>allocs<\/span>), cumulative return (<\/span>cr<\/span>), average daily return (<\/span>adr<\/span>), stan<\/span>da<\/span>rd<\/span> deviation of daily returns (<\/span>sddr<\/span>), and Sharpe ratio (<\/span>sr<\/span>). This project <\/span>builds upon what you learned about portfolio performance metrics and <\/span>optimizers<\/span> to optimize a portfolio. That means that you will find how much of a portfolio\u2019s funds should be allocated to each stock to optimize its performance. While a portfolio can be optimized for many different metrics, in this version of the assignment we will maximize the Sharpe Ratio. You will use the data cleansing techniques introduced in the course lectures.<\/span><\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row _builder_version=”4.4.5″][et_pb_column type=”4_4″ _builder_version=”4.4.5″][et_pb_text _builder_version=”4.9.4″ header_2_font=”|||on|on||||” hover_enabled=”0″ sticky_enabled=”0″]<\/p>\n Implement the optimize_portfolio() function<\/span><\/b>\u00a0<\/span><\/p>\n The optimization.py file must implement this <\/span>API specification<\/span><\/a> (including the default arguments).<\/span>\u00a0<\/span><\/p>\n Revise the code to implement a Python function named optimize_portfolio() in the file optimization.py that can find the optimal allocations for a given set of stocks. You should optimize for maximum Sharpe Ratio taking long positions only.<\/span>\u00a0<\/span><\/p>\n The function should accept as input a list of symbols as well as start and end dates and return a list of floats as a one-dimensional Numpy array that represents the allocations to each of the equities. You can take advantage of routines developed in the optional assess portfolio (<\/span>see note under<\/span><\/i> Starter Code<\/span><\/i>) project to compute daily portfolio value and statistics, and then by using cut-and-paste to move the code from the project for the necessary functions into optimization.py.\u00a0 Otherwise, you will need to create the functions within optimization.py.<\/span>\u00a0<\/span><\/p>\n You are given the following inputs for optimizing a portfolio:<\/span>\u00a0<\/span><\/p>\n Your goal is to find allocations to the symbols that optimize the criterion given above. Assume 252 trading days in a year and a risk-free return of 0.0 per day.<\/span>\u00a0<\/span><\/p>\n You will produce a single chart, as a .png file called \u201cFigure1.png\u201d, comparing the optimal normalized portfolio with SPY using the following portfolio parameters:<\/span>\u00a0<\/span><\/p>\n While four (4) symbols are used in the portfolio associated with Figure 1, the implementation must be robust enough to handle any number of symbols.<\/span>\u00a0<\/span><\/p>\n The chart must be correct, be properly titled, have appropriate axis labels, use consistent axis ranges, and have legends. You should use python\u2019s Matplotlib library.<\/span>\u00a0<\/span><\/p>\n The implementation will be evaluated using 10 test cases that will validate that the sum of the allocations to 1.0 (with a -\/+ 0.2 tolerance), the individual allocations range from 0.0 to 1.0 (with a -\/+ 0.2 tolerance), and that the standard deviation is correct (within 5% of the reference implementation).<\/span>\u00a0<\/span><\/p>\n <\/span><\/p>\n Technical Requirements<\/span><\/b>\u00a0<\/span><\/p>\n Use only the functions provided in the util.py file for reading historical stock data provided in ..\/data folder. Do NOT modify, copy, or move the util.py file or its functions.<\/span>\u00a0<\/span><\/p>\n Your code must run in less than 5 seconds per test case in the university-provided Gradescope environment.<\/span>\u00a0<\/span><\/p>\n A random seed may not be set for this assignment.\u00a0<\/span>\u00a0<\/span><\/p>\n Watermarked Charts (i.e., where the GT Username appears over the lines) can be shared in the designated pinned (e.g., \u201cProject 2- Student Charts\u201d) thread alone. Charts presented in reports or submitted for grading must not contain watermarks.<\/span>\u00a0<\/span><\/p>\n Example<\/span><\/b>\u00a0<\/span><\/p>\n The following example illustrates how the optimize_portfolio function will be called:<\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][et_pb_dmb_code_snippet code=”aW1wb3J0IGRhdGV0aW1lIGFzIGR0IAphbGxvY3MsIGNyLCBhZHIsIHNkZHIsIHNyID0gIAogICAgb3B0aW1pemVfcG9ydGZvbGlvKHNkPWR0LmRhdGV0aW1lKDIwMDgsMSwxKSwgZWQ9ZHQuZGF0ZXRpbWUoMjAwOSwxLDEpLCAgCiAgICBzeW1zPVsnR09PRycsJ0FBUEwnLCdHTEQnLCdYT00nXSwgZ2VuX3Bsb3Q9RmFsc2UpIA==” language=”python” _builder_version=”4.9.4″ _module_preset=”default”]aW1wb3J0IGRhdGV0aW1lIGFzIGR0IAphbGxvY3MsIGNyLCBhZHIsIHNkZHIsIHNyID0gIAogICAgb3B0aW1pemVfcG9ydGZvbGlvKHNkPWR0LmRhdGV0aW1lKDIwMDgsMSwxKSwgZWQ9ZHQuZGF0ZXRpbWUoMjAwOSwxLDEpLCAgCiAgICBzeW1zPVsnR09PRycsJ0FBUEwnLCdHTEQnLCdYT00nXSwgZ2VuX3Bsb3Q9RmFsc2UpIA==[\/et_pb_dmb_code_snippet][et_pb_text _builder_version=”4.9.4″ _module_preset=”default”]<\/p>\n Where the returned output is:<\/span>\u00a0<\/span><\/p>\n The input parameters are:<\/span>\u00a0<\/span><\/p>\n An example chart that you might create for assessing your optimizer.<\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][et_pb_image src=”http:\/\/lucylabs.gatech.edu\/ml4t\/wp-content\/uploads\/2021\/05\/p2examplechart.png” title_text=”p2examplechart” align=”center” _builder_version=”4.9.4″ _module_preset=”default”][\/et_pb_image][et_pb_text _builder_version=”4.9.4″ _module_preset=”default”]<\/p>\n Suggestions<\/span><\/b>\u00a0<\/span><\/p>\n constraints = ({ ‘type’: ‘ineq’, ‘fun’: lambda inputs: 50.0 – np.sum(inputs) })<\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][et_pb_text _builder_version=”4.9.4″ _module_preset=”default”]<\/p>\n The optimization.py file must implement this <\/span>API specification<\/span><\/a>.<\/span>\u00a0<\/span><\/p>\n <\/span><\/p>\n To make it easier to get started on the project and focus on the concepts involved, you will be given a starter framework. This framework assumes you have already set up the <\/span>local environment<\/span><\/a> and <\/span>ML4T Software<\/span><\/a>. The framework for Project 1 can be obtained from:\u00a0\u00a0<\/span>Optimize_Something2021Summer.zip<\/span><\/a>.<\/span>\u00a0<\/span><\/p>\n Extract its contents into the base directory (e.g., ML4T_2021Summer). This will add a new folder called \u201coptimize_something\u201d to the directory structure.<\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][et_pb_image src=”http:\/\/lucylabs.gatech.edu\/ml4t\/wp-content\/uploads\/2021\/05\/optimize_structure.png” title_text=”optimize_structure” align=”center” _builder_version=”4.9.4″ _module_preset=”default”][\/et_pb_image][et_pb_text _builder_version=”4.9.4″ _module_preset=”default”]<\/p>\n Within the optimize_something folder are two files: optimization.py, and grade_optimization.py. You will modify the optimization.py file to implement the necessary functionality for this assignment. You must implement the optimize_portfolio() function. The existing code in the optimization.py file is best thought of as \u201cstub\u201d or \u201cstarter\u201d code that may contain ideas for functions and methods that might be used in your implementation. This file must remain and run from within the optimize_something directory using the following command:<\/span>\u00a0<\/span><\/p>\n [\/et_pb_text][et_pb_dmb_code_snippet code=”UFlUSE9OUEFUSD0uLi86LiBweXRob24gb3B0aW1pemF0aW9uLnB5IA==” _builder_version=”4.9.4″ _module_preset=”default”]UFlUSE9OUEFUSD0uLi86LiBweXRob24gb3B0aW1pemF0aW9uLnB5IA==[\/et_pb_dmb_code_snippet][et_pb_text _builder_version=”4.9.4″ _module_preset=”default”]<\/p>\nOverview<\/h2>\n
About The Project<\/h2>\n
Your Implementation<\/span><\/span>\u00a0<\/span><\/span><\/h2>\n
\n
\n
\n
\n
\n
\n
\n
\n
\n
TECHNICAL REQUIREMENTS<\/h2>\n
Starter Code<\/span><\/b>\u00a0<\/span><\/h3>\n