Oddbean new post about | logout
▲ ▼
Bitman | 1 days ago (raw) | export | reply | flag +4 
 All models are wrong, but some are useful and we need to respect them.

The criticism from those who hate S2F of focusing on supply is wrong because the correlation regime with central bank liquidity is very high (R2 > 95 %) and does not fail to signal demand for the latter asset.

https://image.nostr.build/13aa79007f07428637380075934807431fca2ca7d471aac92bc4b19fe3d6dd3d.jpg
▲ ▼
SHSS | 1 days ago (raw) | root | parent | reply | flag 
 Why doesn’t S2F work for gold?
▲ ▼
caleb | 1 days ago (raw) | root | parent | reply | flag 
 Some Cursory thoughts

The S2f for gold is variable and not strictly decreasing (quite static over time) so I don’t know how good of models can be built on top of it.  

Also the s2f of gold is a function of price/demand so it further complicates things (Low one year, higher the next, then back)- not so with BTC.
▲ ▼
caleb | 1 days ago (raw) | root | parent | reply | flag 
 Strictly increasing*
▲ ▼
SHSS | 1 days ago (raw) | root | parent | reply | flag 
 Does SF2 work for litecoin or bitcoin cash?
▲ ▼
caleb | 1 days ago (raw) | root | parent | reply | flag 
 Supply based models don’t account for demand characteristics, and demand is based on if the coin is a good money.
▲ ▼
SHSS | 23 hours ago (raw) | root | parent | reply | flag 
 So what good is a model that doesn’t account for half of what makes up the price?