[Peticolas]

Thanks for the response, Frank. Yes, it’s common in finance literature to standardize returns by dividing returns by the standard deviation of those returns. That’s pretty similar to the Sharpe ratio except it ignores the risk-free rate.

[Peticolas]

I’ve used Sharpe ratio to assess the risk adjusted returns of past performance. I’ve never read that it was useful as a predictive indicator. Do you know of any articles or books that discuss that, Frank?

[Peticolas]

Thanks Frank. I just wanted to get an intuition for the moves within the strategies. I assume NASDAQ 100 equity piece went to its low volatility portfolio for the same reason.

[Peticolas]

Nevermind. It seems that strategies have historical returns. but portfolios (made up of strategies) do not. I understand. The examples I gave were actually portfolios.

[Peticolas]

Thanks Frank. Although I was on the right date, I may have dome something wrong that changed things.

[Peticolas]

Chyylq, they don’t use a data source that includes delisted stocks. So, yes, their backtesting will suffer from some survivorship bias. This is mitigated somewhat by their walk-forward testing. What I do to evaluate their strategies is go to the very bottom of the page for the strategy or portfolio, and there is a link for the actual historical performance of the portfolio or strategy. As you might expect, those metrics fall short of the backtest results, perhaps because of the survivorship bias, perhaps because of overfitting. Still, the historical numbers look quite good for many of their strategies and portfolios. The only strategy I particularly dislike is their crypto strategy. Since crypto has such crazy runs, I suspect their clients were after them to come up with a crypto strategy, but I would rank it least desirable, at least at this point.

[Peticolas]

Thank you, Frank.

[Peticolas]

And although it’s an old post, the historical results for this strategy are excellent. I follow and code several momentum systems, and this beats mine and others I have seen. People are attracted to the Nasdaq 100 because that’s where the volatility is and you need some volatility in momentum strategies. The big downside in trading it exclusively is only four stocks. I suspect that limitation is a function of the smaller universe. The top 1% of stocks in the NASDAQ 100 is … one stock. In the Russell 2000, the top 1% is 20 stocks.

So I would be reluctant to trade it exclusively.

[Peticolas]

There’s an article in which Vangelis describes LI testing as walk-forward analysis. My question is whether LI uses anchored walk-forward analysis (always starting at the same data point) or rolling walk forward analysis (discarding old data).

[Peticolas]

Thanks Vangelis. Your video shows what I expected, but when I enter my username and password, the application threw two errors (screenshots of which I have reported in emails), and then exits the application. It does not appear to be an issue with credentials. We did solve that problem. Tried the normal thing, rebooted, but it still throws the two errors.

[Peticolas]

Thanks Trevor. If I had been able to get the software to work, I probably would have been on-board, but the installation file throws errors that no one seems interested in helping me with.

[Peticolas]

Thanks Trevor. I appreciate the input. And I agree. It’s a lot of work but you can come up with a model of 10 or 20 parameters that perfectly predicts the past, but they’re generally useless in the future. Simpler is better.

[Peticolas]

I like the idea but implementation might be difficult since the Hedge strategy is incorporated into so many of the riskier strategies. What you would really want would be the correlation of the different unhedged strategies.

[Peticolas]

Thanks for the clarification, Frank. My fault. I missed that.

[Peticolas]

Thanks Frank. That makes sense. Generally better to sell options than to buy. And if the delta on my DITM calls declines, I would see theta decay.

[Peticolas]

Just so you know, I went back and looked at the blog posts and pulled the annual performance for the NASDAQ 100 from those posts. In a couple of years, the January post did not disclose annual results. If that was the case, I took 11 month results from the December post.

The NASDAQ 100 strategy page lists historical performance as:

2023: 29.5%
2022: 4.3%
2021: 56.6%
2020: 104.3%
2019: 31.3%

The blogs disclosed performance for those years as:

2023: 29.5%
2022: 2.8%
2021: 24.6%
2020: 5.8%
2019: 11.5%

I doubt this will be posted but if so, evaluate for yourselves.

[Peticolas]

Thanks Frank. I’m on a trial membership and I’m really impressed with your work here at LogicaInvest. I have worked on my own momentum strategies, since this seems to be the biggest anomaly to CAPM and efficient markets, but I have not achieved your risk-adjusted returns. I’m not familiar with the literature on mean reversion, and that may be my disadvantage.

I would love to try your QT software but I still don’t have a password. My browser keeps me logged into the site so it’s not an issue otherwise, but I can’t use QT without a password. I’ve submitted help requests through the contact us form, and I look forward to getting that issue resolved.

[Peticolas]

I would just add that having 4 stocks is weak in terms of diversification, since those four COULD all be in tech. So if we had something analogous to the tech bubble burst in the early 2000s, you might experience losses far outside the historical losses included in the back testing.

[Peticolas]

Hi. I have a followup on rebalancing. If your strategy/portfolio composition changes, you would clearly rebalance, but what if the components are the same for the next month. Would you rebalance back to the recommended percentages or would you just leave it alone until the composition of the portfolio changes?

[Author]

Posts