Reduce data queries from yfinance, since queries are limited. Debug failing tests for _bollinger_signals.

Author: niklasniedermeier

src/backtest_bay/data/download_data.py

@@ -24,14 +24,6 @@ def _validate_symbol(symbol):
         error_msg = "Symbol must be a non-empty string."
         raise TypeError(error_msg)
 
-    is_symbol_existent = not yf.Ticker(symbol).history(period="1d").empty
-    if not is_symbol_existent:
-        error_msg = (
-            f"Invalid symbol: '{symbol}'. Please provide a valid ticker symbol "
-            "from https://finance.yahoo.com/"
-        )
-        raise ValueError(error_msg)
-
 
 def _validate_interval(interval):
     valid_intervals = {
@@ -73,16 +65,8 @@ def _validate_date_range(start_date, end_date):
 
 def _validate_output(data, symbol, start_date, end_date, interval):
     if data.empty:
-        # Since we already checked for a valid symbol, we check if input dates
-        # fit to historic available data.
-        ticker = yf.Ticker(symbol)
-        hist = ticker.history(period="max")
-        min_date = hist.index.min().strftime("%Y-%m-%d")
-        max_date = hist.index.max().strftime("%Y-%m-%d")
-
         error_msg = (
             f"No data found for {symbol} between {start_date} and {end_date} "
-            f"with interval '{interval}'.\n"
-            f"Available data range for {symbol}: {min_date} to {max_date}."
+            f"with interval '{interval}'."
         )
         raise ValueError(error_msg)

tests/analysis/test_generate_signals.py

@@ -3,26 +3,25 @@
 
 from backtest_bay.analysis.generate_signals import _bollinger_signals
 
-# Tests for _bollinger_signals
-
 
+# Tests for _bollinger_signals
 def test_bollinger_signals_constant_prices():
     prices = pd.Series([100] * 30)
-    signals = _bollinger_signals(prices)
+    signals = _bollinger_signals(prices, window=20, num_std_dev=2)
     expected = pd.Series([0] * 30, index=prices.index)
     pd.testing.assert_series_equal(signals, expected)
 
 
 def test_bollinger_signals_buy_signal():
-    prices = pd.Series([100] * 20 + [90])
-    signals = _bollinger_signals(prices)
+    prices = pd.Series([100] * 20 + [90] + [100])
+    signals = _bollinger_signals(prices, window=20, num_std_dev=2)
     buy_signal = 2
     assert signals.iloc[-1] == buy_signal
 
 
 def test_bollinger_signals_sell_signal():
-    prices = pd.Series([100] * 20 + [110])
-    signals = _bollinger_signals(prices)
+    prices = pd.Series([100] * 20 + [110] + [100])
+    signals = _bollinger_signals(prices, window=20, num_std_dev=2)
     assert signals.iloc[-1] == 1
 
 
@@ -31,3 +30,10 @@ def test_bollinger_signals_window_effect():
     signals_small_window = _bollinger_signals(prices, window=2, num_std_dev=1)
     signals_large_window = _bollinger_signals(prices, window=20, num_std_dev=1)
     assert not signals_small_window.equals(signals_large_window)
+
+
+def test_bollinger_signals_single_price():
+    prices = pd.Series([100])
+    signals = _bollinger_signals(prices, window=20, num_std_dev=2)
+    expected = pd.Series([0], index=prices.index)
+    pd.testing.assert_series_equal(signals, expected)

tests/data/test_download_data.py

@@ -11,20 +11,6 @@
 
 
 # Tests for _validate_symbol
-def test_validate_symbol_valid_cases():
-    _validate_symbol("AAPL")
-
-
-def test_validate_symbol_invalid_invalid_symbol():
-    with pytest.raises(ValueError, match="Invalid symbol: 'INVALID'"):
-        _validate_symbol("INVALID")
-
-
-def test_validate_symbol_invalid_empty_string():
-    with pytest.raises(ValueError, match="Invalid symbol: ''"):
-        _validate_symbol("")
-
-
 def test_validate_symbol_invalid_non_string():
     with pytest.raises(TypeError, match="Symbol must be a non-empty string."):
         _validate_symbol(123)
@@ -131,18 +117,13 @@ def test_validate_output_valid_input():
 def test_validate_output_empty_input():
     # generate typical Yahoo Finance output format
     symbol = "AAPL"
-    arrays = [
-        ["Close", "High", "Low", "Open", "Volume"],
-        [symbol, symbol, symbol, symbol, symbol],
-    ]
-    index = pd.MultiIndex.from_tuples(
-        list(zip(*arrays, strict=False)), names=["Price", "Ticker"]
+    empty_data = pd.DataFrame()
+    start_date = "1800-01-01"
+    end_date = "1800-05-01"
+    interval = "1d"
+    match = (
+        f"No data found for {symbol} between {start_date} and {end_date} "
+        f"with interval '{interval}'."
     )
-
-    empty_data = pd.DataFrame(columns=index)
-
-    empty_data.index.name = "Date"
-
-    match = "No data found for AAPL between 1800-01-01 and 1800-05-01."
     with pytest.raises(ValueError, match=match):
         _validate_output(empty_data, symbol, "1800-01-01", "1800-05-01", "1d")