WilderMA

talipp/indicators/ATR.py

@@ -7,6 +7,45 @@
 from talipp.ohlcv import OHLCV
 
 
+class WilderMA(Indicator):
+    """Wilder's Moving Average.
+
+    Input type: `float`
+
+    Output type: `float`
+
+    Args:
+        period: Period.
+        input_values: List of input values.
+        input_indicator: Input indicator.
+        input_modifier: Input modifier.
+        input_sampling: Input sampling type.
+    """
+
+    def __init__(self, period: int,
+                 input_values: List[float] = None,
+                 input_indicator: Indicator = None,
+                 input_modifier: InputModifierType = None,
+                 input_sampling: SamplingPeriodType = None):
+        super().__init__(input_modifier=input_modifier,
+                         input_sampling=input_sampling)
+
+        self.period = period
+        self.k = 1.0 / self.period
+
+        self.initialize(input_values, input_indicator)
+
+    def _calculate_new_value(self) -> Any:
+        #if not has_valid_values(self.input_values, self.period):
+        if len(self.input_values) < self.period:
+            return None
+        elif has_valid_values(self.input_values, self.period, exact=True):
+        #elif len(self.input_values) == self.period:
+            return sum(self.input_values[-self.period:]) / self.period
+        else:
+            return float(self.k * self.input_values[-1] + (1.0 - self.k) * self.output_values[-1])
+
+
 class ATR(Indicator):
     """Average True Range
 
@@ -35,22 +74,9 @@ def __init__(self, period: int,
         self._tr = TrueRange()
         self.add_sub_indicator(self._tr)
 
-        self.tr = []
-
-        self.add_managed_sequence(self.tr)
+        self._ma_tr = WilderMA(period, input_indicator=self._tr)
 
         self.initialize(input_values, input_indicator)
 
     def _calculate_new_value(self) -> Any:
-        tr = self._tr.output_values[-1]
-        if has_valid_values(self.input_values, 1, exact=True):
-            self.tr.append(tr)
-        else:
-            self.tr.append(tr)
-
-        if len(self.input_values) < self.period:
-            return None
-        elif len(self.input_values) == self.period:
-            return sum(self.tr) / self.period
-        else:
-            return (self.output_values[-1] * (self.period - 1) + self.tr[-1]) / self.period
+        return self._ma_tr.output_values[-1]