Financial Modeling
July 9, 2024
Here, I provide two tools operating at the confluence of statistics and finance. The first is a forecasting tool that uses the Prophet Model to fit and
predict the daily adjusted closing price of a specified asset. You can look up ticker symbols
here. The data we will be using is provided by Yahoo Finance and the Federal Reserve Bank.
The Prophet Model is known for its accuracy and flexibility. However, the volatility of closing prices varies depending on the asset itself, so the responsible user will
invoke their personal judgment and use this merely as a tool to complement (or contradict) their personal claims.
As shown below, you can input the ticker symbol, forecasting period, and time unit. Note, the forecasting period will depend on the asset's maturity.
I limit the start date to 5 years prior in order to help mitigate computational cost and enhance predictive accuracy.
Although you may ask the model to forecast far into the future, I recommend looking at more immediate forecasts using small time intervals such as
days and months.
You will notice, the farther into the future the model predicts, the wider the uncertainty interval is, indicating less reliable predictions.
The second tool is one for portfolio optimization and valuation. This tool computes the efficiency frontier of a portfolio using 1,000 simulations and notes the
optimal portfolio based on the Sharpe Ratio. Once the optimal portfolio is identified,
relevant performance metrics are displayed, such as expected return, volatility, and asset weights.
Moreover, the Security Market Line
is fit and beta coefficients are plotted to portray the valuation of each asset as well as the overall
portfolio. Note, the market return is based on the S&P 500 Index (^GSPC).