What is Tema?
Tema stands for Triple Exponential Moving Average. It is a type of moving average used to smooth a time series (such as a price chart) while trying to reduce the lag that typically comes with smoothing. In plain terms, it aims to follow the underlying direction more closely than a simple exponential moving average.
Tema is commonly discussed within the broader topic of moving averages, where the goal is to transform noisy observations into a clearer line. However, it is still an indicator derived from past data: it does not “know” the future.
How Tema works (mechanics)
A moving average is computed from past values, and exponential moving averages (EMAs) assign more weight to recent observations.
Tema is constructed in three stages using exponential smoothing:
- Smooth the original series once
- Compute an EMA of the input series (for example, closing prices). Call this first smoothed series EMA1.
- Smooth again
- Compute another EMA, but this time of EMA1. Call this EMA2.
- Smooth a third time
- Compute EMA3 by applying the EMA again to EMA2.
Then Tema combines these components to reduce lag. A commonly used form is:
- Tema = 3 × EMA1 − 3 × EMA2 + EMA3
This structure is designed so that the most recent behavior is reflected more strongly, while the older, lagging effects are partially canceled out by the combination.
Key inputs
- Smoothing period (often written as a length): controls how quickly Tema responds to new changes.
- Input data: the series you smooth (typically price). Different data choices can change the appearance and behavior of the output.
How to interpret Tema in practice
Tema produces a single smoothed series that you can compare to the underlying data.
Typical interpretation is qualitative:
- When Tema turns and moves in the same direction as price behavior, it may suggest that the smoothed trend is shifting.
- When Tema flattens or whipsaws, it may indicate that the market is not moving in a clean direction.
Because Tema is still derived from past values, it can appear late after abrupt reversals. Lag reduction helps, but it cannot eliminate delay completely.
Limitations and risks
1) Parameter sensitivity
The smoothing period strongly affects behavior:
- Shorter periods usually make Tema react faster, but they also tend to increase sensitivity to short-term noise.
- Longer periods usually smooth more and reduce noise, but the line can become less responsive.
This means that two users applying the same concept with different parameter values can see materially different results.
2) Lag is reduced, not removed
Tema’s construction is intended to reduce lag compared with a basic EMA, but it remains a function of historical data. After sudden changes, any moving average can still lag because it aggregates information over time.
3) Choppy or range conditions can degrade usefulness
In sideways or highly variable conditions, smoothing techniques can create oscillations and frequent changes in direction. Even with lag reduction, Tema can “flip” repeatedly when the underlying series alternates up and down.
4) Data choices and computation details matter
Different conventions (for example, which price field is used, how the first EMA values are initialized, and how missing data is handled) can change the resulting series. This makes exact comparisons between implementations difficult unless you match the computation details.
5) Overfitting during evaluation
If you test Tema using many combinations of parameter settings and timeframes until you find a pattern that looks good, you may fit noise rather than a stable relationship. The risk is not unique to Tema, but it becomes relevant whenever a method is tuned extensively.
Verification and independent assessment
For concepts like Tema, verification is about testing whether the behavior you observe is consistent across time and conditions.
Reasonable approaches include:
- Evaluate multiple periods to see whether the effect is stable.
- Check performance across different market regimes (trending versus sideways) rather than relying on one window.
- Use out-of-sample evaluation to reduce the risk of tailoring to the past.
Importantly, any evaluation should focus on understanding how Tema behaves and what tradeoffs you accept (responsiveness versus noise), not on assuming a specific outcome.
Comparison within moving averages (what makes Tema distinct)
Compared with a single exponential moving average, Tema uses repeated smoothing plus a corrective combination. That extra structure is the main distinction:
- A basic EMA performs one smoothing step.
- Tema applies multiple smoothing steps and then combines them to counteract some lag.
This design generally aims for a faster-moving average line without fully sacrificing smoothness.
If you need related context, you can also read about moving averages generally to understand how smoothing and lag trade off across common variants.