91Ó°ÊÓ

Probability is a fundamental concept in statistics that measures how likely an event is to occur. In our coin-flipping exercise, each coin flip is an independent event, meaning that the result of one flip does not affect the outcome of another.
The theoretical probability of getting heads on any given flip is 0.5, because there are only two possible outcomes: heads or tails. This is known as the classical definition of probability, which assumes all outcomes are equally likely. However, determining probability from observed outcomes, known as **experimental probability**, can sometimes show variations, especially in small samples. When interpreting probability, it's important to consider whether the sample size is large enough to accurately represent the true likelihood of each outcome.

• Probability of heads: 0.5 (theoretical)
• Probability from experiment: can vary, especially in small samples
• Result from friend's experiment: 0.7

The Law of Large Numbers is a key principle in probability theory and statistics. This law states that as the number of trials in an experiment increases, the experimental probability of an outcome will tend to approach the theoretical probability.
For example, if you continuously flip a fair coin, the proportion of heads is likely to get closer to 0.5 as more flips are conducted. In the scenario with your friend, using only 10 coin flips is not sufficient for this convergence to occur. This is because with fewer trials, each individual outcome has a larger impact on the overall experimental probability.
So, the apparent discrepancy between the observed probability (0.7) and the theoretical probability (0.5) isn't surprising with only 10 flips. If he continued flipping the coin a hundred, a thousand, or even more times, the outcome percentages would likely stabilize closer to the expected probability.

• Small numbers = more fluctuation
• Large numbers = closer to theoretical probability

Sample size is a crucial factor in determining the reliability of experimental results. A small sample size, like the 10 flips in the exercise, can lead to misinterpretations because it doesn't capture enough data to witness the average outcome reliably.
When increasing the sample size, you progressively reduce the impact of each individual event's outcome on the overall result. This means that in statistics, the bigger the sample, the more reliable your data in approximating the true probabilities. Your friend's conclusion about the coin not being balanced is premature because it was based on a sample size too small to represent the true nature of random flips effectively.
To gain a more accurate understanding of the coin's fairness, significantly more flips are required. This way, the conjecture aligns better with the theoretical probability, reducing anomalies caused by chance variations in smaller samples.

• Small sample size: prone to higher variability
• Larger sample size: reduces variability and approximates theoretical probability

Experimental probability is the probability computed from the actual outcomes of an experiment. It is calculated by dividing the number of times an event occurs by the total number of trials conducted. In the exercise with the coin flips, the experimental probability was calculated as 7 heads out of 10 flips, equating to 0.7.
One must remember that experimental probability can deviate significantly from theoretical probability, particularly when the number of trials is limited. Over time, as experiments are repeated and the Law of Large Numbers takes effect, these two types of probabilities will align more closely.
To improve the reliability of experimental probability, it is paramount to ensure the number of trials conducted is sufficiently large. This reduces the skew caused by any individual anomaly, ensuring a more balanced and representative sample of outcomes. Hence, advising your friend to increase the number of flips will illustrate the coin's probability more accurately, reflecting the true conditions of randomness and fairness.

• Based on actual outcomes
• Prone to variation with few trials
• Aligns closer to theoretical probability with more trials