"I agree with you, but the theoretical point still remains :
Given a chart, is it not legitimate to expect that its predictive ability, if any, should not only extent to one day in the future (ie the right hand side of the chart) but also one day in the past (ie one day to the left of the start of the chart) ?"
An important consequence of the above is this:
If prices in one period shape prices in other periods, then it should be possible for a chartist to tell whether a chart with no labels is the right way up.
No ?
Add to My Watchlist
What is My Watchlist?