Does anybody have a Scan for the Kaufman’s Efficiency Ratio?

brobin526

I have KAMA but would like to scan by just the ER. If somebody knows a way to do that, would be much appreciated. Thank you.

markd

I don't know much about KAMA, but after reading the chart school article, it seems you might be able to use "1" for the smoothing constants to get the ER value (e.g. (10,1,1).

You could try plotting the KAMA on a few charts with those values (make a special list), note the current values for KAMA on the charts, then scan that list for a range of KAMA values ( e.g., > 1 and < 3 - just making up those values) that includes some charts and excludes others, and see if it picks the right ones. Do that after the market has closed so the values don't change.

brobin526

great thank you for the info.

lmkwin

If I understand it correctly Efficiency Ratio is absolute value net change for the period divided by the sum of the individual absolute values of net change that make up the period.

From the ChartSchool Article on KAMA:
ER = Change/Volatility

Change = ABS(Close - Close (10 periods ago))

Volatility = Sum10(ABS(Close - Prior Close))

Volatility is the sum of the absolute value of the last ten price changes (Close - Prior Close).

In theory the below (example for EV(5)) "could" work but there isn't a good way to test it, that I know of. Also when you get into "complex" equations, the results may be suspect. It passed syntax but I wouldn't rely on it without a good way to test it.

[AbsVal(ROC(5)) / [AbsVal(ROC(1)) + 1 days ago AbsVal(ROC(1)) + 2 days ago AbsVal(ROC(1)) + 3 days ago AbsVal(ROC(1)) + 4 days ago AbsVal(ROC(1))] > 0]

markd

I'm assuming you can rely on the built-in KAMA(x,y,z) function to calculate the ER with the "x" parameter correctly. If you use 1 for the "y" and "z" parameters, I think you are effectively multiplying by 1 because they reduce to (1+1) squared = 1, which would yield the ER.

lmkwin

TOS has an Efficiency Ratio indicator. Basically ER is designed to show trending or not trending. ER will rise in an up trend or down trend. Similar to the ADX line, but a bit more sensitive. ER values are between 0 and 1.

KAMA is a moving average that utilizes the ER but isn't valid source for the ER.

KAMA x,y,z, x is the ER period, the y and z are smoothing EMA lines of x. The faster ma crosses below the slower ma the KAMA line may fall. Faster ma above slower ma, the KAMA line may rise.

Results are plotted as a moving average on the price chart with the Price of the security being a factor in the plotted line.

A scan looking for KAMA 10,1,1 > 99 will show securities that have a KAMA line greater than 99. The ER difference doesn't effectively factor into the results other than showing some securities with a Price less than KAMA value and some with a Price greater than the KAMA value requested in the scan. The securities may or may not be in a trend so it's not filtering ER.

ER is the AbsVal of the total period x change/ AbsVal of the individual periods change for the total period x. ER is most similar in computation to the Chande Momentum Oscillator, also available on TOS.

If you want to use KAMA 10,1,1 for example, it would probably make sense to use Slope KAMA 10,1,1 or compare values KAMA 10,1,1 versus KAMA 10,1,1 x periods ago. Slope > or < a certain value may discern if ER is in trending or not trending. KAMA > or < x periods ago KAMA by a certain limit could also indicate a higher efficiency ratio. Higher ER is just a trend indicator. Up Trend or Down Trend

I think there are probably easier ways to scan trend and trend momentum, like PPO, RSI, ADX, Price Channels, or pretty much any other indicator available in the SC technical indicators dropdown, but that's just my opinion.