Add crossover indicator

Author: MDUYN

README.md

@@ -14,12 +14,15 @@ pip install pyindicators
 
 * Native Python implementation, no external dependencies needed except for Polars or Pandas
 * Dataframe first approach, with support for both pandas dataframes and polars dataframes
-* Trend indicators
+* Trend indicators(#trend-indicators)
   * [Simple Moving Average (SMA)](#simple-moving-average-sma)
   * [Exponential Moving Average (EMA)](#exponential-moving-average-ema)
-* Momentum indicators
+* Momentum indicators(#momentum-indicators)
   * [Relative Strength Index (RSI)](#relative-strength-index-rsi)
   * [Relative Strength Index Wilders method (RSI)](#wilders-relative-strength-index-wilders-rsi)
+* Indicator helpers(#indicator-helpers)
+  * [Crossover](#crossover)
+  * [Is Crossover](#is-crossover)
 
 ## Indicators
 
@@ -129,6 +132,47 @@ pd_df.tail(10)
 
 ### Indicator helpers
 
+#### Crossover
+
+```python
+from polars import DataFrame as plDataFrame
+from pandas import DataFrame as pdDataFrame
+
+from investing_algorithm_framework import CSVOHLCVMarketDataSource
+from pyindicators import crossover, ema
+
+# For this example the investing algorithm framework is used for dataframe creation,
+csv_path = "./tests/test_data/OHLCV_BTC-EUR_BINANCE_15m_2023-12-01:00:00_2023-12-25:00:00.csv"
+data_source = CSVOHLCVMarketDataSource(csv_file_path=csv_path)
+
+pl_df = data_source.get_data()
+pd_df = data_source.get_data(pandas=True)
+
+# Calculate EMA and crossover for Polars DataFrame
+pl_df = ema(pl_df, source_column="Close", period=200, result_column="EMA_200")
+pl_df = ema(pl_df, source_column="Close", period=50, result_column="EMA_50")
+pl_df = crossover(
+    pl_df,
+    first_column="EMA_50",
+    second_column="EMA_200",
+    result_column="Crossover_EMA"
+)
+pl_df.show(10)
+
+# Calculate EMA and crossover for Pandas DataFrame
+pd_df = ema(pd_df, source_column="Close", period=200, result_column="EMA_200")
+pl_df = ema(pd_df, source_column="Close", period=50, result_column="EMA_50")
+pd_df = crossover(
+    pd_df,
+    first_column="EMA_50",
+    second_column="EMA_200",
+    result_column="Crossover_EMA"
+)
+pd_df.tail(10)
+```
+
+![CROSSOVER](https://github.com/coding-kitties/PyIndicators/blob/main/static/images/indicators/crossover.png)
+
 #### Is Crossover
 
 ```python

pyindicators/__init__.py

@@ -1,9 +1,12 @@
-from .indicators import sma, rsi, is_crossover, crossunder, ema, wilders_rsi
+from .indicators import sma, rsi, is_crossover, crossunder, ema, wilders_rsi, \
+    crossover, is_crossover
 
 __all__ = [
     'sma',
     'is_crossover',
     'crossunder',
+    'crossover',
+    'is_crossover',
     'ema',
     'rsi',
     "wilders_rsi"

pyindicators/indicators/__init__.py

@@ -1,12 +1,13 @@
 from .simple_moving_average import sma
-from .crossover import is_crossover
+from .crossover import is_crossover, crossover
 from .crossunder import crossunder
 from .exponential_moving_average import ema
 from .rsi import rsi, wilders_rsi
 
 __all__ = [
     'sma',
     'is_crossover',
+    "crossover",
     'crossunder',
     'ema',
     'rsi',

pyindicators/indicators/crossover.py

@@ -1,6 +1,65 @@
 from typing import Union
+
 from pandas import DataFrame as PdDataFrame
 from polars import DataFrame as PlDataFrame
+import polars as pl
+
+
+def crossover(
+    data: Union[PdDataFrame, PlDataFrame],
+    first_column: str,
+    second_column: str,
+    result_column = "crossover",
+    data_points: int = None,
+    strict: bool = True,
+) -> Union[PdDataFrame, PlDataFrame]:
+    """
+    Identifies crossover points where `first_column` crosses above or below `second_column`.
+
+    Args:
+        data: Pandas or Polars DataFrame
+        first_column: Name of the first column
+        second_column: Name of the second column
+        result_column (optional): Name of the column to
+            store the crossover points
+        data_points (optional): Number of recent rows to consider (optional)
+        strict (optional): If True, requires exact crossovers; otherwise,
+            detects when one surpasses the other.
+
+    Returns:
+        A DataFrame with crossover points marked.
+    """
+
+    # Restrict data to the last `data_points` rows if specified
+    if data_points is not None:
+        data = data.tail(data_points) if isinstance(data, PdDataFrame) else data.slice(-data_points)
+
+    # Pandas Implementation
+    if isinstance(data, PdDataFrame):
+        col1, col2 = data[first_column], data[second_column]
+        prev_col1, prev_col2 = col1.shift(1), col2.shift(1)
+
+        if strict:
+            crossover_mask = ((prev_col1 < prev_col2) & (col1 > col2)) | ((prev_col1 > prev_col2) & (col1 < col2))
+        else:
+            crossover_mask = (col1 > col2) | (col1 < col2)
+
+        data[result_column] = crossover_mask.astype(int)
+
+    # Polars Implementation
+    elif isinstance(data, PlDataFrame):
+        col1, col2 = data[first_column], data[second_column]
+        prev_col1, prev_col2 = col1.shift(1), col2.shift(1)
+
+        if strict:
+            crossover_mask = ((prev_col1 < prev_col2) & (col1 > col2)) | ((prev_col1 > prev_col2) & (col1 < col2))
+        else:
+            crossover_mask = (col1 > col2) | (col1 < col2)
+
+        # Convert boolean mask to 1s and 0s
+        data = data.with_columns(pl.when(crossover_mask).then(1).otherwise(0).alias(result_column))
+
+    return data
 
 
 def is_crossover(