Add alternate election challenge solution

somacdivad · somacdivad · commit dd4561978168 · 2019-11-15T09:37:54.000-06:00
diff --git a/ch08-conditional-logic/9-challenge-simulate-an-election.py b/ch08-conditional-logic/9-challenge-simulate-an-election.py
diff --git a/ch08-conditional-logic/9a-challenge-simulate-an-election.py b/ch08-conditional-logic/9a-challenge-simulate-an-election.py
@@ -0,0 +1,42 @@
+# 8.9 - Challenge: Simulate an Election
+# Solution to challenge
+
+
+# Simulate the results of an election using a Monte Carlo simulation
+
+from random import random
+
+num_times_A_wins = 0
+num_times_B_wins = 0
+
+num_trials = 10_000
+for trial in range(0,num_trials):
+    votes_for_A = 0
+    votes_for_B = 0
+
+    # Determine who wins the 1st region
+    if random() < 0.87:
+        votes_for_A = votes_for_A + 1
+    else:
+        votes_for_B = votes_for_B + 1
+
+    # Determine who wins the 2nd region
+    if random() < 0.65:
+        votes_for_A = votes_for_A + 1
+    else:
+        votes_for_B = votes_for_B + 1
+
+    # Determine who wins the erd region
+    if random() < 0.17:
+        votes_for_A = votes_for_A + 1
+    else:
+        votes_for_B = votes_for_B + 1
+
+    # Determine overall election outcome
+    if votes_for_A > votes_for_B:
+        num_times_A_wins = num_times_A_wins + 1
+    else:
+        num_times_B_wins = num_times_B_wins + 1
+
+print(f"Probability A wins: {num_times_A_wins / num_trials}")
+print(f"Probability B wins: {num_times_B_wins / num_trials}")
diff --git a/ch08-conditional-logic/9b-challenge-simulate-an-election.py b/ch08-conditional-logic/9b-challenge-simulate-an-election.py
@@ -0,0 +1,59 @@
+# 8.9 - Challenge: Simulate an Election
+# Alternate solution to challenge
+
+
+# Simulate the results of an election using a Monte Carlo simulation
+
+from random import random
+
+
+def run_regional_election(chance_A_wins):
+    """Return the result of a regional election, either "A" or "B".
+
+    The chances of "A" winning are determined by chance_A_wins.
+    """
+    if random() < chance_A_wins:
+        return "A"
+    else:
+        return "B"
+
+
+def run_election(regional_chances):
+    """Return the result of an election, either "A" or "B".
+
+    regional_chances is a list or tuple of floats representing the
+    chances that candidate "A" will win in each region.
+
+    For example, run_election([.2, .5, .7]) will run an election with
+    three regions, where candidate "A" has a 20% chance to win in the
+    first region, 50% in the second, and 70% in the third.
+    """
+    num_regions_won_by_A = 0
+    for chance_A_wins in regional_chances:
+        if run_regional_election(chance_A_wins) == "A":
+            num_regions_won_by_A = num_regions_won_by_A + 1
+
+    # Return the results. Note that the number of regions won by candidate
+    # "B" is the total number of regions minus the number of regions won by
+    # candidate "A". The total number of regions is the same as the length
+    # of the regional_chances list.
+    if num_regions_won_by_A > len(regional_chances) - num_regions_won_by_A:
+        return "A"
+    else:
+        return "B"
+
+
+CHANCES_A_WINS_BY_REGION = [0.87, 0.65, 0.17]
+NUM_TRIALS = 10_000
+
+# Run the Monte-Carlo simulation
+num_times_A_wins = 0
+for trial in range(NUM_TRIALS):
+    if run_election(CHANCES_A_WINS_BY_REGION) == "A":
+        num_times_A_wins = num_times_A_wins + 1
+
+# Display the probabilities that candidate A or candidate B wins the
+# election. Note the probability that B wins can be calculated by
+# subtracting the probability that A wins from 1.
+print(f"Probability A wins: {num_times_A_wins / num_trials}")
+print(f"Probability B wins: {1 - (num_times_A_wins / num_trials)}")
