Correct expected profit (#43)

* Update .md files * Update examples
rgaveiga · Jan 26, 2025 · 217c7cd · 217c7cd
1 parent f05b0cc
commit 217c7cd
Show file tree

Hide file tree

Showing 13 changed files with 62 additions and 116 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,9 @@
 # CHANGELOG
 
+## 1.4.2 (2025-01-25)
+
+- Removed `expected_profit` and `expected_loss` calculation from `_get_pop_bs` in support.py; implementation was not correct, giving wrong results when compared with Monte Carlo simulations
+
 ## 1.4.1 (2025-01-04)
 
 - Removed a small bug in `create_price_seq` in support.py

diff --git a/README.md b/README.md
@@ -8,14 +8,12 @@ a user-defined target date, the range of stock prices for which the strategy is
 profitable (i.e., generating a return greater than \$0.01), the Greeks associated with 
 each leg of the strategy using the Black-Sholes model, the resulting debit or credit on the 
 trading account, the maximum and minimum returns within a specified lower and higher price 
-range of the underlying asset, and an estimate of the strategy's probability of profit, 
-expected profit and expected loss.
+range of the underlying asset, and an estimate of the strategy's probability of profit.
 
-The probability of profit (PoP), expected profit and expected loss at the user-defined target 
-date for the strategy are calculated by default using the Black-Scholes model. Alternatively,
-the user can provide an array of underlying asset prices following a distribution other than 
-the normal (e.g. Laplace) or model other than the Black-Scholes model (e.g. Heston model) that 
-will be used in the calculations.
+The probability of profit (PoP) on the user-defined target date for the strategy is calculated 
+by default using the Black-Scholes model. Alternatively, the user can provide an array of 
+underlying asset prices following a distribution other than the normal (e.g. Laplace) or 
+model other than the Black-Scholes model (e.g. Heston model) that will be used in the calculations.
 
 Despite the code having been developed with option strategies in mind, it can also be 
 used for strategies that combine options with stocks and/or take into account the 
@@ -91,7 +89,8 @@ Jupyter notebooks in the **examples** directory.
 Contributions are definitely welcome. However, it should be mentioned that this 
 repository uses [poetry](https://python-poetry.org/) as a package manager and 
 [git hooks](https://git-scm.com/book/en/v2/Customizing-Git-Git-Hooks) with 
-[pre-commit](https://pre-commit.com/) to customize actions on the repository.
+[pre-commit](https://pre-commit.com/) to customize actions on the repository. Source 
+code must be formatted using [black](https://github.com/psf/black).
 
 ## Disclaimer
 

diff --git a/examples/black_scholes_calculator.ipynb b/examples/black_scholes_calculator.ipynb
@@ -43,7 +43,7 @@
      "output_type": "stream",
      "text": [
       "Python version: 3.11.9 | packaged by Anaconda, Inc. | (main, Apr 19 2024, 16:40:41) [MSC v.1916 64 bit (AMD64)]\n",
-      "OptionLab version: 1.4.1\n"
+      "OptionLab version: 1.4.2\n"
      ]
     }
    ],
@@ -106,7 +106,7 @@
      "output_type": "stream",
      "text": [
       "CPU times: total: 0 ns\n",
-      "Wall time: 5.82 ms\n"
+      "Wall time: 0 ns\n"
      ]
     }
    ],

diff --git a/examples/calendar_spread.ipynb b/examples/calendar_spread.ipynb
diff --git a/examples/call_spread.ipynb b/examples/call_spread.ipynb
diff --git a/examples/covered_call.ipynb b/examples/covered_call.ipynb
@@ -54,7 +54,7 @@
      "output_type": "stream",
      "text": [
       "Python version: 3.11.9 | packaged by Anaconda, Inc. | (main, Apr 19 2024, 16:40:41) [MSC v.1916 64 bit (AMD64)]\n",
-      "OptionLab version: 1.4.1\n"
+      "OptionLab version: 1.4.2\n"
      ]
     }
    ],
@@ -119,8 +119,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "CPU times: total: 46.9 ms\n",
-      "Wall time: 52.4 ms\n"
+      "CPU times: total: 203 ms\n",
+      "Wall time: 205 ms\n"
      ]
     }
    ],
@@ -142,7 +142,7 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.legend.Legend at 0x225b149b510>"
+       "<matplotlib.legend.Legend at 0x2e55fd93910>"
       ]
      },
      "execution_count": 5,
@@ -195,8 +195,6 @@
      "text": [
       "Probability of profit: 0.522421406650333\n",
       "Profit ranges: [(162.9, inf)]\n",
-      "Expected profit: 1028.0\n",
-      "Expected loss: -919.0\n",
       "Per leg cost: [-16404.0, 114.99999999999999]\n",
       "Strategy cost: -16289.0\n",
       "Minimum return in the domain: -8087.0\n",

diff --git a/examples/naked_call.ipynb b/examples/naked_call.ipynb
diff --git a/log b/log
diff --git a/optionlab/__init__.py b/optionlab/__init__.py
@@ -1,7 +1,7 @@
 import typing
 
 
-VERSION = "1.4.1"
+VERSION = "1.4.2"
 
 
 if typing.TYPE_CHECKING:

diff --git a/optionlab/models.py b/optionlab/models.py
@@ -482,7 +482,7 @@ class Outputs(BaseModel):
     def __str__(self):
         s = ""
 
-        for key, value in self.dict(
+        for key, value in self.model_dump(
             exclude={"data", "inputs"},
             exclude_none=True,
             exclude_defaults=True,

diff --git a/optionlab/support.py b/optionlab/support.py
@@ -5,7 +5,7 @@
 from typing import Optional
 
 import numpy as np
-from numpy import abs, round, arange, sum, exp
+from numpy import abs, round, arange
 from numpy.lib.scimath import log, sqrt
 from scipy import stats
 
@@ -379,9 +379,6 @@ def _get_pop_bs(
         target.
     """
 
-    probability_of_reaching_target = 0.0
-    probability_of_missing_target = 0.0
-
     expected_return_above_target = None
     expected_return_below_target = None
 
@@ -392,52 +389,26 @@ def _get_pop_bs(
     )
 
     for i, t in enumerate(profit_range):
-        prob = []
-        exp_profit = []
-
-        if t == [(0.0, 0.0)]:
-            continue
-
-        for p_range in t:
-            lval = log(p_range[0]) if p_range[0] > 0.0 else -float("inf")
-            hval = log(p_range[1])
-            drift = (
-                inputs.interest_rate
-                - inputs.dividend_yield
-                - 0.5 * inputs.volatility * inputs.volatility
-            ) * inputs.years_to_target_date
-            m = log(inputs.stock_price) + drift
-            w = stats.norm.cdf((hval - m) / sigma) - stats.norm.cdf((lval - m) / sigma)
-
-            if w > 0.0:
-                v = stats.norm.pdf((hval - m) / sigma) - stats.norm.pdf(
+        prob = 0.0
+
+        if t != [(0.0, 0.0)]:
+            for p_range in t:
+                lval = log(p_range[0]) if p_range[0] > 0.0 else -float("inf")
+                hval = log(p_range[1])
+                drift = (
+                    inputs.interest_rate
+                    - inputs.dividend_yield
+                    - 0.5 * inputs.volatility * inputs.volatility
+                ) * inputs.years_to_target_date
+                m = log(inputs.stock_price) + drift
+                prob += stats.norm.cdf((hval - m) / sigma) - stats.norm.cdf(
                     (lval - m) / sigma
                 )
-                exp_stock = round(
-                    exp(m - sigma * v / w), 2
-                )  # Using inverse Mills ratio
-
-                if exp_stock > 0.0 and exp_stock <= s.max():
-                    exp_stock_index = np.where(s == exp_stock)[0][0]
-                    exp_profit.append(profit[exp_stock_index])
-                    prob.append(w)
-
-        if len(t) > 0:
-            prob_array = np.array(prob)
-            exp_profit_array = np.array(exp_profit)
-
-            if i == 0:
-                probability_of_reaching_target = sum(prob_array)
-                expected_return_above_target = round(
-                    sum(exp_profit_array * prob_array) / probability_of_reaching_target,
-                    2,
-                )
-            else:
-                probability_of_missing_target = sum(prob_array)
-                expected_return_below_target = round(
-                    sum(exp_profit_array * prob_array) / probability_of_missing_target,
-                    2,
-                )
+
+        if i == 0:
+            probability_of_reaching_target = prob
+        else:
+            probability_of_missing_target = prob
 
     return (
         probability_of_reaching_target,

diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "optionlab"
-version = "1.4.1"
+version = "1.4.2"
 description = "Python library for evaluating options trading strategies"
 authors = ["Roberto Gomes, PhD <[email protected]>"]
 readme = "README.md"

diff --git a/tests/test_core.py b/tests/test_core.py
@@ -9,8 +9,6 @@
 COVERED_CALL_RESULT = {
     "probability_of_profit": 0.5472008423945267,
     "profit_ranges": [(164.9, float("inf"))],
-    "expected_profit": 2011.0,
-    "expected_loss": -1758.0,
     "per_leg_cost": [-16899.0, 409.99999999999994],
     "strategy_cost": -16489.0,
     "minimum_return_in_the_domain": -9590.000000000002,
@@ -27,7 +25,6 @@
 PROB_100_ITM_RESULT = {
     "probability_of_profit": 1.0,
     "profit_ranges": [(0.0, float("inf"))],
-    "expected_profit": 524.0,
     "per_leg_cost": [-750.0, 990.0],
     "strategy_cost": 240.0,
     "minimum_return_in_the_domain": 240.0,
@@ -44,8 +41,6 @@
 PROB_NAKED_CALL = {
     "probability_of_profit": 0.8389215512144531,
     "profit_ranges": [(0.0, 176.14)],
-    "expected_profit": 115.0,
-    "expected_loss": -707.0,
     "per_leg_cost": [114.99999999999999],
     "strategy_cost": 114.99999999999999,
     "minimum_return_in_the_domain": -6991.999999999999,
@@ -187,8 +182,6 @@ def test_covered_call_w_prev_position(nvidia):
     ) == {
         "probability_of_profit": 0.7048129541301169,
         "profit_ranges": [(154.9, float("inf"))],
-        "expected_profit": 2566.0,
-        "expected_loss": -1390.0,
         "per_leg_cost": [-15899.0, 409.99999999999994],
         "strategy_cost": -15489.0,
         "minimum_return_in_the_domain": -8590.000000000002,
@@ -311,8 +304,6 @@ def test_3_legs(nvidia):
     ) == {
         "probability_of_profit": 0.6790581742719213,
         "profit_ranges": [(156.6, float("inf"))],
-        "expected_profit": 2997.0,
-        "expected_loss": -1447.0,
         "per_leg_cost": [-15899.0, -750.0, 990.0],
         "strategy_cost": -15659.0,
         "minimum_return_in_the_domain": -8760.000000000002,
@@ -488,8 +479,6 @@ def test_calendar_spread():
     ) == {
         "probability_of_profit": 0.599111819020198,
         "profit_ranges": [(118.87, 136.15)],
-        "expected_profit": 2960.0,
-        "expected_loss": -835.99,
         "per_leg_cost": [4600.0, -5900.0],
         "strategy_cost": -1300.0,
         "minimum_return_in_the_domain": -1300.0000000000146,