See also: trading by "reading the chart"...
Show this thread
Noticing a pattern isn't quite how you predict the future. A random event can follow a pattern -- until it doesn't. If there's no likely causal explanation that ties your pattern to the event, then it's not a reliable model.
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Explains superstition
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
correlation doesn't necessarily mean causation I guess?
@fcholletThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You need some time series hands on. Simple features. Stocks prices might be a bad start for you because your lack of practical skills working with non elliptical distributions. I love this subject . What is the problem you are trying to tackle (eventually) ?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Which is precisely why self-driving based on deep learning is condemned to failure. cc
@missy_cummings@Tweetermeyer@filippie509@lizadixonThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Precisely
-
And most systems are non-linear dynamical systems, which means they can't be predicted anyway
- Show replies
New conversation -
-
-
Sun rises, sun sets. Sun rises, sun sun sets. Sun rises, ...I predict a sunset, even without a causal theory.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The problem of induction?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Sure but don’t we usually use them when we are not certain of exact causal relation? In which case potentially unreliable predictor would still be better to have than no predictor? Or are there other options?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.