Quote from: d5000 on April 16, 2025, 03:55:01 PM

Picking a single event is even more cherry picking,

Of course it is cherry picking because my thesis is that what we call correlations are nothing that exist in reality, they are human projections. For you, from 2020 onwards, there will be observed regularities of which my cherry picking would be an exception (I have chosen the most colorful “de-correlation”) and for me there are no such correlations, they are independent events that in some cases coincide temporally.

Quote from: d5000 on April 16, 2025, 03:55:01 PM

and if you had read what I wrote above in other posts, I don't question that there are price movements which are caused by reasons internal to the Bitcoin ecosystem, which of course often lead to divergence with the stock markets.

Yes I had read it and in the same way I could say that I have not denied that in the short term there are similarities, as for example a Fed announcement can push both the S&P 500 and bitcoin price in one direction or the other.

Quote from: d5000 on April 16, 2025, 03:55:01 PM

Let's look more closely to the post-2020 movements and correlations:

This is a clear case of two people looking at the same thing and seeing completely different things.

Quote from: d5000 on April 16, 2025, 03:55:01 PM

In the periods marked by green circles, there is almost always a positive correlation (the zero-correlation line was "reforced" by me as it wasn't visible) between S&P 500 and Bitcoin. The red circles mark periods of significant negative correlations. The May 2021 event you mentioned is here also quite well visible. And in the rest of the periods I don't see a clear tendency.

We can see that there are long stretches of positive correlations which last several months with only very small "dips" into zero or negative correlation territory. 2023 is the only year since 2020 without a clear positive correlation.

Okay, my thesis was that what we call correlation only exists in the short term and that if we zoom out there is no such correlation. Then you say you do see it since 2020, which is taking a partial sample and we could say cherry picking as well. Within that 5 year sample there are according to you a majority of positive correlations and a few other “negative” correlations which in my words would be a sign that there is no correlation.

I think this is a conceptual thing. You start from the assumption that there is a correlation and your way of analyzing data will confirm your initial concept, in that sense we can say that it is biased. And in the same way I assume that there is no such correlation and we can also say that I am biased in reverse.

Quote from: d5000 on April 16, 2025, 03:55:01 PM

For me, the long "correlated" stretches are very unlikely to be a coincidence. But I don't think there is a direct causality, as explained above. It's more related to independent variables like Fed interest rates, general market sentiment etc..

Lol. It just wondered after all the discussion if we are simply saying the same thing in different ways.

Grin