91Ó°ÊÓ

Probability is a fundamental concept in statistics that measures the likelihood of a certain event occurring. In the context of our circuit board problem, we deal with a specific type of probability called binomial probability. It is applicable when there are only two possible outcomes, such as success or failure, for each trial.

For example, the probability that a single diode will fail is given as 0.01, or 1%. This is a simple scenario where the diode can either fail (which happens with 1% probability) or work (which happens with 99% probability).

• **Independent Trials**: Each diode’s operation is independent. The success of one does not affect another.
• **Fixed Number of Trials**: With 200 diodes, the probability calculations consider all diodes together.

These aspects make the situation a perfect match for binomial probability, allowing for calculations using formulas specific to the binomial distribution.

Standard deviation is a measure of how spread out the outcomes are in a probability distribution. In the binomial context, it tells us how much variation to expect from the mean number of events (like failing diodes in circuit boards).

To find the standard deviation of diodes failing, you can use the formula:\[\sigma = \sqrt{np(1-p)}\]where \( n \) is the total number of diodes (200), and \( p \) is the probability of failure (0.01). This results in a standard deviation of approximately 1.41.

• **Interpretation**: A small standard deviation means most diode failures will be close to the mean of 2.

By understanding this concept, you gain insight into how predictable the number of failing diodes is on any given board.

Normal approximation is a technique used to make binomial probability calculations easier. When you have a large number of trials, like 200 diodes, and the probability of success is small, the binomial distribution resembles a normal distribution.

We apply this method by calculating the z-score which represents how far a certain raw score (like the number of failing diodes) lies from the mean, adjusted for standard deviation:\[ z = \frac{x - \mu}{\sigma}\]where \( x \) is your value of interest (such as 3.5), \( \mu \) is the mean, and \( \sigma \) is the standard deviation.

• **Why 3.5?**: When calculating probabilities like "at least 4 failures," a continuity correction (next topic) slightly shifts this to accommodate a discrete-to-continuous distribution transition.

This approximation simplifies complex binomial probability calculations, making them more manageable using normal distribution tables or software.

Continuity correction is an adjustment technique when applying normal approximation to a discrete distribution like the binomial distribution. Since the binomial outcomes (number of diode failures) are discrete, while the normal distribution is continuous, a small correction helps align the two.

For example, if you want the probability of "at least four diodes failing," you would technically look for \( P(X \geq 4) \). Using a continuity correction, you adjust this to \( P(X > 3.5) \), because it smooths the binomial step-function to better fit the normal curve.

• **Adjustment Explained**: This correction involves decimals to bridge the gap between whole numbers (diodes) and the continuous nature of the normal distribution, improving accuracy.

By implementing continuity correction, the approximation aligns more closely with the actual probabilities, especially when binomial and normal distributions overlap.