This is a trend-following tool based on the Nadaraya-Watson (NW) kernel regression estimator.
Nadaraya-Watson Estimator: This indicator employs a weighted average of historical prices where the weights are determined by a kernel function. This kernel function gives more weight to recent prices and less weight to older ones. The kernel function's shape (Gaussian, Epanechnikov, or a combination) can be chosen by the user, and its bandwidth controls the sensitivity of the estimator. A smaller value means more sensitivity to short-term changes; a larger value means smoother curves and less sensitivity.
Trend Confirmation: The indicator also uses a confirmation period (confirm_period) to minimize whipsaws. The trend only changes once the estimator has moved in a particular direction for a configurable number of consecutive bars.
Bandwidth (h): This is the key parameter controlling the estimator's sensitivity to price changes.
Smaller values (e.g., 10-40) result in a more reactive and wiggly line. This will be more sensitive to short-term price changes.
Larger values (e.g., 100-200) result in a smoother line, focusing on longer-term trends.
Source: The price data used in the calculations. The default is close but can be changed to other price values such as high, low, or a mathematical combination.
Kernel Type: Select the type of kernel function to use.
Confirmation Period: Adjust this setting to specify how many consecutive bars the estimator must be moving in the same direction before the trend is confirmed.
Lower values result in more trend changes.
Higher values result in fewer trend changes and can filter out some noise.
Ce script est publié en code source fermé et vous pouvez l'utiliser librement. Vous pouvez le préférer pour l'utiliser sur un graphique. Vous ne pouvez pas visualiser ou modifier son code source.
Les informations et les publications ne sont pas destinées à être, et ne constituent pas, des conseils ou des recommandations en matière de finance, d'investissement, de trading ou d'autres types de conseils fournis ou approuvés par TradingView. Pour en savoir plus, consultez les Conditions d'utilisation.