The risk scores are just a way to estimate how risky an investment is. They are a simplified view of what's actually happening. In that sense, they could be calculated in any way that is plausible and acceptable to whoever is interested in solving the problem. For example: from a conversation with experts, from past volatility of the investments (though we all know past behavior isn't a good indicator of future behavior), from forecasts on the investments' performance, etc. In more complicated portfolio optimization models, this simple idea of a risk score is replaced by some measure of the variance of the returns of the portfolio, which turns the model into a non-linear model.

@@TallysYunes Thank you so much for answer and explanation. I'm clear in how the linear program function.

You're welcome!

In this video I already give you the complete mathematical model and just go over it briefly. In class, I build it from scratch. I'm planning on making another collection of videos in which I create all the models from scratch, but they're not available yet. This is on my to-do list for 2022.

You're welcome. Glad to hear the video was helpful. Constraint (b) is just an example of how to implement a weighted average type of constraint. You weigh each investment by how risky it is and average them out. This is a very simplified measure of risk. In practice, people use more complicated (and non-linear) formulas for that. But since this is an early example in the course, I'm keeping it simple for now.

@@TallysYunes Dr.Yunes, Thank you so much for taking the time to reply. It’s very much appreciated. Thank you for offering informative explanation. I will definitely keep studying your lectures to learn more. Have a wonderful day, Kindest regards, Mad

Thank you so much!

I'm not sure what you mean.