CD2H gitForager

IBM/Portfolio-compute

Name: Portfolio-compute

Owner: International Business Machines

Description: Analyze an existing Investment Portfolio using IBM Cloud services

Created: 2017-11-08 17:35:59.0

Updated: 2018-05-02 22:14:07.0

Pushed: 2018-05-17 21:22:10.0

Homepage: https://developer.ibm.com/code/patterns/compute-analytics-on-an-investment-portfolio

Size: 815

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

Portfolio Analytics Computation

Upload a portfolio and compute analytics on it

This journey leverages two services:

The Investment Portfolio service is used to manage and store portfolios and financial security information such as the set of eligible investments, benchmarks, and user portfolios.
The Instruments Analytics service is used to compute analytics on securities. This service is pre-integrated with market data and leverages a wide set of financial models. These financial models are cashflow generation models, meaning they project all cash flow a given security is going to make over its lifespan in order to calculate properties of financial securities such as current value (using present value of the cash flows) or sensitivities (to see how much cashflows change when incremental changes are made to various model inputs).

This code pattern is designed for developers with interest in creating financial applications pertaining to investment portfolios. When the reader has completed this journey, they will understand how to:

Load and retrieve data from the Investment Portfolio service
Construct a payload to generate analytics on a portfolio
Construct a user interface to display results, or download to a CSV file.

Flow

User uploads a portfolio for analysis.
User selects a portfolio and which analytic to compute on the portfolio.
The portfolio and analytic choices are sent to the Instrument Analytics service. Computational results are returned to the browser or downloaded in CSV format.

Included Components

Note: these services are free for those who have a Lite account

Steps

Use the Deploy to IBM Cloud button OR create the services and run Run Locally.

Deploy to IBM Cloud

Create an IBM Cloud account and directly deploy the application using the button below.

Running the Application Locally

Follow these steps to setup and run this developer journey. The steps are described in detail below.

Prerequisites

1. Clone the repo

Clone the Portfolio Analytics Computation code locally. In a terminal, run:

$ git clone https://github.com/IBM/Portfolio-compute.git

2. Create IBM Cloud services

Create the following services in IBM Cloud. These services are part of either Free or Experimental plan.

3. Configure Manifest file

Edit the manifest.yml file in the folder that contains your code and replace with a unique name for your application. The name that you specify determines the application's URL, such as your-application-name.mybluemix.net. Additionally - update the service names so they match what you have in the IBM Cloud. The relevant portion of the manifest.yml file looks like the following:

clared-services:
Investment-Portfolio:
  label: fss-portfolio-service
  plan: fss-portfolio-service-free-plan
Instrument-Analytics:
  label: fss-instrument-analytics-service
  plan: fss-instrument-analytics-service-free-plan
plications:
path: .
memory: 128M
instances: 1
name: Portfolio-Compute
disk_quota: 1024M
domain: mybluemix.net
services:
- Investment-Portfolio
- Instrument-Analytics
buildpack: python_buildpack

4. Configure .env file

Create a .env file in the root directory of your clone of the project repository by copying the sample .env.example file using the following command in terminal:

 .env.example .env

NOTE Most files systems regard files with a “.” at the front as hidden files. If you are on a Windows system, you should be able to use either GitBash or Xcopy

You will need to update the credentials with the IBM Cloud credentials for each of the services you created in Step 2.

The .env file will look something like the following:

vestment Portfolio
_PORTFOLIO_USERID_W=
_PORTFOLIO_PWD_W=
_PORTFOLIO_USERID_R=
_PORTFOLIO_PWD_R=

strument Analytics
ri=https://fss-analytics.mybluemix.net/
ccess_token=

5. Run Application

In your terminal, cd into this project's root directory

Run pip install -r requirements.txt to install the app's dependencies
Run python run.py
Access the running app in a browser at http://0.0.0.0:8080/

You can push the app to IBM Cloud using IBM Cloud CLI. This will use the services and application name in the manifest.yml file. From your root directory login into IBM Cloud using CLI:

ogin

And push the app to IBM Cloud:

ush

6. Upload Holdings

Once the application is running, the first step is to upload a file that will be used to create a portfolio or a series of portfolios in the Investment Portfolio service. We use the file format of the Algorithmics Risk Service (ARS) import file as many production clients are already used to that format. You can find an example file in this repo labelled “Blue Chip Portfolio”. Alternatively, the file ICFS_SAMPLE_POSITIONS.csv contains the superset of instruments that the Instrument Analytics service supports.

The column labeled “UNIQUE ID” must refer to the unique identifier of the asset in our system.
The “NAME” column will hold the display name of the asset.
“POSITION UNITS” column holds the quantity.
“PORTFOLIO” indicates which portfolio the asset belongs to.

The code will create a portfolio for each unique element found in the “PORTFOLIO” column. Future releases of this code will take into account a portfolio hierarchy, but currently each portfolio is entirely independent of each other.

Some notes:

The portfolio will be loaded as 500-asset chunks as there are currently limitations on POST request sizing. This means you shouldn't use the 'latest=True' parameter when requesting calculations from the Instrument Analytics service!
The portfolio will be tagged as type = 'unit test portfolio' to distinguish between any other portfolios that may exist in the system.

Note: You can navigate to /api/unit_test_delete to delete all portfolios that have been loaded into the service using this application. Specifically, it looks for portfolios with “type = unit test portfolio”. You can always start over this way, but be careful to not access this if you've done considerable work.

7. Compute Analytics

Once the portfolio have been loaded, a second API call can be made. This was parsed out as two separate calls as uploading the portfolio may not need to occur frequently, whereas the computations may.

The next step is to use the application to call the /api/unit_test endpoint. This will perform the following:

Gather all portfolios in the Investment Portfolio service that are labelled as type = 'unit test portfolio'
Parse each portfolio into 500-asset chunks to be sent to the Instrument Analytics service (currently a limitation we enforce)
Compute a series of analytics, which is currently hard-coded to 'THEO/Price' and 'THEO/Value'. This can be changed in the code.
Return a csv file of results, along with printing out statistics on timing to the console.

Some notes:

This script currently processes securities in series (synchronously). As the Instrument Analytics service runs on Kubernetes, this script can be enhanced to submit calculation requests asynchronously to improve timing.
Instruments not found will be ignored.

Troubleshooting

To troubleshoot your IBM Cloud application, use the logs. To see the logs, run:

ogs <application-name> --recent

If you are running locally - inspect your environment variables closely to confirm they match. Try running each service as standalone:

on InvestmentPortfolio.py
on InstrumentAnalytics.py

License

Apache 2.0

This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.