Home > Technical analysis > Indicators and oscillators > Triple Exponential Moving Average (TEMA, or TRIX)
Triple Exponential Moving Average (TEMA, or TRIX)
The Triple Exponential Moving Average (TEMA) is a unique combination of a single exponential moving average, a double exponential moving average, and a triple exponential moving average that provides less lag than any of those three individually. It is not just a moving average of a moving average of a moving average. It can be used instead of traditional moving averages for smoothing price data or other indicators.
As well as a momentum indicator, TRIX (the Triple Exponential Moving Average) is an oscillator, which follows overbought and oversold markets. For that watch a negative value demonstrate an oversold market and positive value to demonstrate an overbought market. When TRIX is used as a momentum indicator, a negative value suggests momentum is decreasing while a positive value suggests increasing momentum. Some analysts think that the TRIX crossing above the zero line is a purchase signal and a closing below the zero line is a sell signal. Distinction between price and TRIX can also show important market turning points.

TRIX filtration of market noise and its tendency to be a leading and not a lagging indicator are two advantages of TRIX over other trend indicators. Insignificant cycles are excluded by using triple exponential smoothing. It is able to lead a market as it calculates the distinction between each bar's smoothed version of the price information. TRIX is best used together with another market-timing indicator so as to reduce false signals when used as a leading indicator.
An exponential moving average of the data is taken for the given period for calculating TRIX. Afterwards an exponential moving average is taken of that result for the same period, followed by another for the second result. The percent change in value of the third moving average is then returned as the value of the TRIX. The value of the TRIX at the beginning of a data series is considered to be zero. Since it uses exponential moving averages, its primary values comprise in its calculation the zero value. It is possible to ignore values before 3 times the period has finished.