Syntax
public TrendQ(DataSeries ds, int period1, int period2, int m, int n, double c, bool useRMSNoise, string description)
public static TrendQ Series(DataSeries ds, int period1, int period2, int m, int n, double c, bool useRMSNoise)
Parameter Description
| DataSeries |
Typically the Close DataSeries reference |
| period1 |
EMA fast period used for the CPC (cumulative price change) indicator. |
| period2 |
EMA slow period used for the CPC indicator. |
| m |
Period for the piecewise EMA of cumulative price change from the slow/fast CPC EMA crossovers. |
| n |
Period used for noise calculation. Generally you should specify n >= m |
| c |
Correction factor for the signal (trend) to noise ratio. (Greatly affects the amplitude of TrendQ.) |
| useRMSNoise |
Pass true to use an RMS noise calculation, otherwise false for linear noise. |
Description
The Trend Quality (Q-indicator) from the April 2004 issue of
Stocks & Commodities magazine determine trends
"with a procedure that attempts to recognize, in a timely way, promising and nonpromising (sic) price trends within the extracted semicycles as well as estimate their strength".
The Q-indicator is a centered oscillator that fluctuates around a zero line with no upper or lower limits. TrendQ measures trend quality. Per the author, David Sepiashvili, readings can be interpreted as follows:
- -1 to +1: trend buried beneath noise
- -1 to -2 or +1 to +2: weak trending
- -2 to -5 or +2 to +5: moderate trending
- below -5 or above +5: strong trending
Calculation
(See open source)
Example
(Please request)