FE Mathematics: Probability and Statistics Review

Probability and statistics form the backbone of data-driven engineering decisions. From assessing the reliability of a system to analyzing experimental results, these tools allow you to quantify uncertainty, make predictions, and validate designs against empirical evidence. Mastering these concepts is not just about passing the FE exam; it's about building a foundational skill set for a competent engineering career.

Foundational Probability and Counting

Probability measures the likelihood of an event occurring, expressed as a value between 0 (impossible) and 1 (certain). The core rules are the building blocks for more complex analysis. The probability of an event $A$ is $P(A)$. The complement rule states $P(A^{\prime })=1-P(A)$. For two events, the addition rule is $P(A\cup B)=P(A)+P(B)-P(A\cap B)$. For independent events, where one does not affect the other, the multiplication rule is $P(A\cap B)=P(A)\times P(B)$.

Before calculating probabilities, you often need to count possible outcomes. Permutations count the number of ways to arrange $r$ items from a set of $n$ distinct items, where order matters: $$P(n,r)=\frac{n!}{(n-r)!}$$. Combinations count the number of ways to choose $r$ items from $n$, where order does not matter: $$C(n,r)=\left(\genfrac{}{}{0ex}{}{n}{r}\right)=\frac{n!}{r!(n-r)!}$$. On the exam, use your calculator's nPr and nCr functions. A key strategy: if the problem mentions "teams," "committees," or "samples," it's likely a combination. If it mentions "lineups," "codes," or "seating arrangements," it's likely a permutation.

Descriptive Statistics and Data Summaries

Descriptive statistics summarize the main features of a dataset. The mean ($\mu$ for population, $\bar{x}$ for sample) is the arithmetic average. The median is the middle value when data is sorted, and the mode is the most frequent value. For the FE exam, you must understand measures of spread. Variance is the average of the squared deviations from the mean. For a population: $${\sigma }^2=\frac{\sum (x_i-\mu {)}^2}{N}$$. For a sample: $$s^2=\frac{\sum (x_i-\bar{x}{)}^2}{n-1}$$. The standard deviation is the square root of the variance ($\sigma$ or $s$), providing a measure of spread in the original data units.

Your calculator's statistics mode is your fastest tool. Input a data list to instantly compute the sample mean ($\bar{x}$), sample standard deviation ($s_x$), and population standard deviation (${\sigma }_x$). Always check whether a problem describes a population or a sample to choose the correct standard deviation.

Key Probability Distributions

Probability distributions describe how probabilities are distributed over the values of a random variable. Discrete distributions apply to countable outcomes. The binomial distribution models the number of successes $k$ in $n$ independent trials, each with success probability $p$. The probability is given by: $$P(X=k)=\left(\genfrac{}{}{0ex}{}{n}{k}\right)p^k(1-p{)}^{n-k}$$. Use it for pass/fail, defect/no-defect scenarios. The Poisson distribution models the number of events occurring in a fixed interval of time or space, with a known average rate $\lambda$: $$P(X=k)=\frac{{\lambda }^k{e}^{-\lambda }}{k!}$$. Use it for modeling rare events like system failures per year.

The most important continuous distribution is the normal (Gaussian) distribution, characterized by its bell-shaped curve, mean $\mu$, and standard deviation $\sigma$. Probabilities are found by converting a value $x$ to a z-score: $z=\frac{x-\mu }{\sigma }$, which measures how many standard deviations $x$ is from the mean. You then use the standard normal table (or calculator function) to find the area under the curve. The empirical rule (68-95-99.7 rule) provides quick estimates: about 68% of data falls within $\pm 1\sigma$ of $\mu$, 95% within $\pm 2\sigma$, and 99.7% within $\pm 3\sigma$.

On the FE exam, use your calculator's normalcdf (or equivalent) function to find probabilities directly without z-table lookup. For inverse problems (finding the data value for a given percentile), use the invNorm function.

Statistical Inference: Confidence Intervals & Hypothesis Testing

Statistical inference allows you to draw conclusions about a population based on a sample. A confidence interval provides a range of plausible values for a population parameter (like a mean or proportion). For a population mean with a large sample (or known population standard deviation), the interval is: $$\bar{x}\pm z_{\alpha /2}\frac{\sigma }{\sqrt{n}}$$. The critical value $z_{\alpha /2}$ corresponds to your confidence level (e.g., 1.96 for 95% confidence). The term $\frac{\sigma }{\sqrt{n}}$ is the standard error of the mean. The interpretation: you are, for example, 95% confident that the true population mean lies within the calculated interval.

Hypothesis testing is a formal method for testing a claim about a population. You start with two hypotheses: the null hypothesis ($H_0$) states a status quo or no-effect claim (e.g., $\mu ={\mu }_0$). The alternative hypothesis ($H_a$ or $H_1$) states what you are trying to evidence (e.g., $\mu \ne {\mu }_0$, $\mu >{\mu }_0$, or $\mu <{\mu }_0$). The test proceeds by calculating a test statistic (like a z-score or t-score) from the sample data and comparing it to a critical value or, more commonly on the FE, calculating a p-value. The p-value is the probability of obtaining a sample result at least as extreme as the one observed, assuming $H_0$ is true. If the p-value is less than the chosen significance level $\alpha$ (commonly 0.05), you reject $H_0$.

Your calculator can perform these tests (Z-Test, T-Test) directly. Input the sample statistics, hypothesized mean, and the form of $H_a$. The output will provide the test statistic and the p-value, which you simply compare to $\alpha$.

Linear Regression and Correlation

Linear regression models the linear relationship between an independent (explanatory) variable $x$ and a dependent (response) variable $y$. The model is $y={\beta }_0+{\beta }_{1}x+ϵ$, where ${\beta }_0$ is the y-intercept, ${\beta }_1$ is the slope, and $ϵ$ represents error. The least-squares method finds the line that minimizes the sum of the squared vertical distances between the data points and the line.

The strength and direction of the linear relationship is measured by the correlation coefficient, $r$, which ranges from -1 to +1. $|r|$ close to 1 indicates a strong linear relationship; $r$ close to 0 indicates a weak one. The coefficient of determination, $r^2$, represents the proportion of variance in $y$ that is explained by the linear model with $x$.

On the exam, use your calculator's linear regression function (LinReg). Input paired $(x,y)$ data. The output will give you ${\beta }_0$ (often labeled a), ${\beta }_1$ (often labeled b), and $r$ and $r^2$. You may be asked to use the equation for prediction: $\hat{y}=b_0+b_{1}x$.

Common Pitfalls

1. Confusing Permutations and Combinations: The most frequent counting error. Remember: order matters = permutation; order doesn't matter = combination. If you are selecting a president, vice-president, and treasurer from 10 people, it's a permutation (P(10,3)) because the roles are distinct. If you are simply selecting a 3-person committee, it's a combination (C(10,3)).
2. Misapplying the Normal Distribution: The normal distribution is continuous. A common trap is using it to find $P(X=k)$ for a discrete problem, which is always zero for a true continuous variable. For binomial problems with large $n$, you may approximate with the normal, but you must apply a continuity correction (e.g., finding $P(X\le 10)$ becomes $P(X<10.5)$ when using the normal approximation).
3. Misinterpreting Confidence Intervals: A 95% confidence interval does not mean "there is a 95% probability that the population mean is in this interval." The population mean is fixed, not random. The correct interpretation is about the method: if you repeated the sampling process many times, 95% of the constructed intervals would contain the true mean.
4. Confusing Significance with Practical Importance: A statistically significant result (small p-value) merely indicates the observed effect is unlikely due to chance alone. It does not mean the effect is large or practically important for an engineering application. Always consider the magnitude of the effect in its real-world context.

Summary

• Probability & Counting: Master the addition, multiplication, and complement rules. Use nPr for permutations (order matters) and nCr for combinations (order doesn't matter) on your calculator.
• Distributions: Know when to apply Binomial (fixed trials, independent, success/failure), Poisson (events in an interval), and Normal (continuous, bell-shaped data). Use normalcdf and invNorm for fast normal distribution solutions.
• Descriptive Statistics: Calculate sample mean ($\bar{x}$) and standard deviation ($s_x$) directly using your calculator's statistics mode. Understand the difference between a sample and a population statistic.
• Statistical Inference: Construct confidence intervals for means and perform hypothesis tests by comparing the p-value (from Z-Test/T-Test) to the significance level $\alpha$. Correctly interpret both in context.
• Linear Regression: Use LinReg to find the least-squares line, correlation coefficient $r$, and $r^2$. The slope $b_1$ represents the predicted change in $y$ for a one-unit increase in $x$.