Portfolio Optimization Models: Performance Data and Metrics

The dataset provided contains detailed financial and performance data for five different models over a period of five years. The dataset includes key metrics such as <b>Sharpe Ratio</b>, <b>Expected Return</b>, <b>Risk (Standard Deviation)</b>, and performance metrics (<b>Performance Start</b> and <b>Performance End</b>) for each model, as well as simulated values across multiple years. Below is a description of the dataset:Dataset Structure:<b>Model</b>: The type of portfolio optimization model being evaluated. The dataset contains five different models:<b>Proposed MTR-SAFER</b><b>Markowitz Model</b><b>MAD Model</b><b>Deep Learning Model</b><b>Reinforcement Learning Model</b><b>Year</b>: The dataset spans five years, with data for each model evaluated across each of these years.<b>Sharpe Ratio</b>: A key metric used to measure the risk-adjusted return of each model. It is calculated by dividing the expected return by the risk (standard deviation). The values are randomly simulated between 0.5 and 1.0 to represent the performance over the years.<b>Expected Return</b>: The predicted return for each model in each year. It is randomly simulated within the range of 0.05 (5%) to 0.15 (15%) to indicate varying returns over time.<b>Risk (Standard Deviation)</b>: The risk or volatility associated with each model's returns. This is represented by the standard deviation of the model’s return and is randomly simulated within the range of 0.03 (3%) to 0.1 (10%).<b>Performance Start</b>: The starting performance value for each model at the beginning of each year, randomly simulated between 0.8 and 1.0. This could represent the initial value of a portfolio managed by the model.<b>Performance End</b>: The final performance value for each model at the end of the year, simulated between 1.0 and 1.1, representing the portfolio's value at the end of the year.