portfolize.

This is an index-driven portfolio modeling software that will serve as your guideline for optimizing your returns!

Comment 2

portfolize

Start Investing Today!

Does investing scare you? Honestly, it scares me too! This is why we, Brian and I, have developed a simple web-app that strategizes an index-driven investment portfolio, one of the most stable forms of investment, for all of us! Whether you like to actively trade or hold on to your money until you see your account flooding on returns, you will be able to strategize your very own portfolio, taylored for your own goals. No, we don't advertize cliche big-name stocks on you. Instead, we simply provide you with the most optimal index based portfolio of instruments that you can choose to invest in!

How Do You Determine Which Indices to Invest In?

We have referenced to Goldman Sach's portfolio strategy that includes: US Large Cap Markets, Fixed Income Derivatives, US Real Estate, Emerging Market Debt, and even some international markets such as International Real Estate and Private Equity! We have gathered the historical data on indices for different types of markets to create the most optimal portfolio. Additionally, we have also taken into account the Sharpe Ratio, a risk adjusted variable, to include the effect of volatility in our model.

What Languages/Packages Did You Guys Use?

For developing the model, we primarily used Python's pandas and statsmodels data analysis tool to analyze and forecast the future expected prices of indices using time series regression and AutoRegression Modeling, a simple machine learning method. For front-end, we decided to build this on a webapp platform for a universal access. As expected, we've used html, CSS and flask mostly for the front end. We've also deployed the website using Heroku platform!

What Challenges Did You Guys Run Into, and What Did You Guys Learn?

Trying to come up with the most accurate and efficient modeling was definitely the most challenging task. Deploying the website was also very difficult but we managed to do it. I definitely learned a lot about predictive modeling, especially time series regression and it was a good experience for Brian to polish his front-end development and deployment skills!

What's next for portfolize

We really want to come up with a better model for predicting future index prediction. Also, we would like to stabilize the model and make it outlier-proof.

Built With

Share this project:

Updates