ETF Analysis and Optimization with Python

Learn how to analyze and optimize an ETF portfolio using Python! In this step-by-step tutorial, we explore a diverse range of ETFs, calculate key performance metrics like returns, volatility, and cumulative returns, and then optimize the portfolio using the Sharpe Ratio. Whether you're a Python enthusiast or a finance professional, this video will guide you through the essentials of portfolio analysis and optimization.

🔗 Check out my website for more Python for Finance:

💻 Get the Notebook/Source Code:
Unlock all resources by becoming a Tier-3 Channel Member here:

What You'll Learn:
How to calculate ETF performance metrics (returns, volatility, Sharpe Ratio).
How to optimize an ETF portfolio using Python.

Optimize your ETF portfolio and take your Python finance skills to the next level! 🚀

[anilmm]

Excellent analysis algovibes you really are an inspiration please keep continuing

[scientifictraders]

Thank you for sharing your expertise in applying modern portfolio theory in pandas. Super math! I'll assume that your incorporate the risk-free return into this public blueprint in your courses. But as caveat to others, the risk-free rate is significant currently, whereas it was just the opposite a few years ago.

[simondechoisy]

I would be fascinated to see this combined with a momentum strategy. I would like to suggest that the strategy would have a list of ETFs which would be ranked based on their 90 day exp linear regression slope x R2(coefficient of determination) as long as the stock is above it's 100 day moving average. The strategy would each buy the top n stocks each month and then hold them for a month. Before running the process again. The optimisation would use the technique outlined in you above video based on 20 trading day volatility. I think this would be a fantastic video and would be of great interest.

[justcars]

It's really good to trade indices and ETFs, they are way less noisy, and clear, they also respect a lot of trading theories. Which make it easy to be profitable trading them.

[Kig_Ama]

I would not make the combination based on historical data, but rather based on the forward earnings for different regions that determine ETF shares

[michal.nalevanko]

Vielen Dank für dieses großartige Video! 🎉 Ich fand den Schritt-für-Schritt-Ansatz und die Erklärung zur Portfolio-Optimierung echt genial. Super hilfreich für Finanz-Nerds wie mich! 🙌 Macht weiter so!

[ThePumaX]

Danke shön! Very useful for a starter like me

[beograd]

Have you tried to optimize over a longer period of time (e.g. 7 years which you could use as training data) and with those optimal weights to calculate the next year returns?

Training horizon: 2016-2023
Test horizon: 2024

You would include more stability and a clearer picture of optimal weights.
You can observe, that the weights based on one trading year are instable and thus showing a relativley poor performance. (According to the comparison of the sharpe ratio and realized returns).

But however. I really like your video as it tries to incorporate portfolio theory on ETFs. And it was funny to see you struggle to decide whether to explain the inverse weights variance matrix or to just go on with the video 😁

[santievangelio]

I got this error:ValueError: shapes (41, ) and (40, ) not aligned: 41 (dim 0) != 40 (dim 0). The line that creates this is: portfolio_return=np.dot(weights, mean_returns).

[RobertGerami]

y finance was previworking, but as of this year, it fails to download any data. I have seen a large number of posts about that issue. So far, nothing I have tried works.