Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : ajtj's Post-Lobotomy Market Charts and Thoughts -- Ignore unavailable to you. Want to Upgrade?

To: ajtj99 who wrote (96225)	10/25/2025 1:26:41 AM
From: Sun Tzu	1 Recommendation Recommended By ajtj99 Read Replies (2) \| Respond to of 96601

Thank you! I needed that. BTW, some of the stories you tell of your wife are very scary. I admire your resolve, both of yours, for doing this good work. But as a friend, I suggest that you both give serious thought to a change of career.

To: ajtj99 who wrote (96225)	10/25/2025 1:41:36 AM
From: Sun Tzu	1 Recommendation Recommended By ajtj99 Read Replies (1) \| Respond to of 96601

I was staring at the charts and I started articulating this idea I have had for an alternative form of regression channels for analyzing stocks and commodities. The concept has been composting in my head for a few years now. I keep coming back to it every several months or so. I assigned the mastery of algorithmic trading and quantitative research to ChatGPT and asked it to keep poking holes into my theory. After nearly 1.5 hours of debate it agreed that I am right and that most quants throw away information that is actually material to detecting regime change. The problem is that although I have a fair idea of how it should work, I still haven't worked out how a machine could do it. There are a lot of things that human can do looking at a chart and a machine can't because we just "get it". We absorb the chart in its entirety. Nobody has to define for you what the trend is or what constitutes a peak or a trough. But making a machine see it is not easy. There are a lot of edge cases and provisos that make it tedious. And as I have explained before, I don't code things that require too many moving parts. So I am not going to start curve fitting to parameterize when a trough or trend change is significant and when it is not; there won't be anything like "if a slope changes more than M and lasts longer than T, then it is significant." I need to clean the logic, not the parameters. And usually that is not easy.