Trendiness of Days - assigning a value
http://www.elitetrader.com/vb/showthread.php?s=&threadid=24779&highlight=trendiness
http://www.elitetrader.com/vb/showthread.php?s=&threadid=24779&highlight=trendiness
How would you test for trendiness?
- Thread starter jerryz
- Start date
Besides tenderness, personality as well as money management is often also important.
http://www.elitetrader.com/vb/showthread.php?s=&postid=932335#post932335
Surely all you have is the respective value of a "move ", as a measure of the total inherent value of the thing being measured, compared to another.
All markets will display the same behaviour as another, at some point in time.
Did you mean, like the concept of "currencies trending well" compared to other markets, that kind of thing?
Inherent behaviour of a single chart, over time-or current timetables of a chart, VS another, now?
What a bunch of foolish answers. There are different was to "measure" a trend but most depend on a fixed lookback # of bars to determine the "trendiness". I've used several but discarded most in favor of a *very simple* momentum-based system designed to catch every trend. If the system is profitable and the equity curve goes up in a smooth slope, then I rate that equity/index as favorable to a trend-based approach. Most indicators like ADX just trick you into thinking there is a trend but fail to tell you if you can trade it profitably!
i am thinking along these lines...
the best trends are the ones that
1) steadily go in one direction
2) have very few pullbacks, and
3) when there are pullbacks, they are small
how would you go about measuring this? what statistical measure would you use?
then i'm thinking...
say you have 2 markets. one market is much more volatile than the other. within a certain time frame, the more volatile market should show more trends. however, this doesn't necessarily mean that the more volatile market is more trendy than the less volatile one. it's just that it's more volatile. within a certain time frame, you will also see more "fake trends" and pullbacks in the more volatile market.
so then i'm thinking...
how do you account for volatility when measuring trendiness? what statistical measure would you use? when a market goes up say 10 days in a row, how can you measure if that's just an expected occurence given the volatility, or if the market is truly trending?
do you guys agree/disagree? any ideas? thanks.
the best trends are the ones that
1) steadily go in one direction
2) have very few pullbacks, and
3) when there are pullbacks, they are small
how would you go about measuring this? what statistical measure would you use?
then i'm thinking...
say you have 2 markets. one market is much more volatile than the other. within a certain time frame, the more volatile market should show more trends. however, this doesn't necessarily mean that the more volatile market is more trendy than the less volatile one. it's just that it's more volatile. within a certain time frame, you will also see more "fake trends" and pullbacks in the more volatile market.
so then i'm thinking...
how do you account for volatility when measuring trendiness? what statistical measure would you use? when a market goes up say 10 days in a row, how can you measure if that's just an expected occurence given the volatility, or if the market is truly trending?
do you guys agree/disagree? any ideas? thanks.
I'm not sure if this is what you're looking for:
FER = "Fractal Efficiency Ratio" coined by Perry Kaufman
FER = (total change in price over a given period) /
(sum of the absolute values of all of the daily price changes)
also:
EI = "Efficiency Index" coined by Van Tharp
EI = (difference between closing prices of today and X periods ago) /
(ATR over the same period)
Tharp defines 2 EI periods: short term and long term
i.e. EI_20 = (C - Ref(C,-20))) / SMA (ATR(1),20)
C is current bar price, Ref(C,-20) is 20 bars ago price
AvgEI_short_term = (EI_20 + EI_45) / 2
AvgEI_long_term = (EI_20 + EI_45 + EI_90 + EI_180) / 4
An efficient stock has AvgEI > 0 and may be as high as 15.
Van Tharp publishes lists of most efficient uptrend / downtrend efficient stocks in his weekly newsletter.
FER = "Fractal Efficiency Ratio" coined by Perry Kaufman
FER = (total change in price over a given period) /
(sum of the absolute values of all of the daily price changes)
also:
EI = "Efficiency Index" coined by Van Tharp
EI = (difference between closing prices of today and X periods ago) /
(ATR over the same period)
Tharp defines 2 EI periods: short term and long term
i.e. EI_20 = (C - Ref(C,-20))) / SMA (ATR(1),20)
C is current bar price, Ref(C,-20) is 20 bars ago price
AvgEI_short_term = (EI_20 + EI_45) / 2
AvgEI_long_term = (EI_20 + EI_45 + EI_90 + EI_180) / 4
An efficient stock has AvgEI > 0 and may be as high as 15.
Van Tharp publishes lists of most efficient uptrend / downtrend efficient stocks in his weekly newsletter.
Hey,
Say you have a set of prices, how would you define and test for profit-potential?
thanks. this is interesting, same kind of idea as the sharpe ratio. the higher the number, the smoother the trend.
however, say we have a market that moves like this:
moves up 20 days in a row, then
moves down 20 days in a row, then
moves up 20 days in a row, then
moves down 20 days in a row.
the ending price is now the same as the starting price. (assuming same increment each day)
in this case, both of the ratios would be 0. but from a trading perspective, this market is great. 4 nicely predicable trends. the two ratios won't capture this.
are there statistical measures that would measure this?