Random Variables, Sampling Distributions, Confidence Intervals, and Hypothesis Testing: Study Guide

Random Variables and Probability Distributions

Definition and Types of Random Variables

A random variable is a numeric value associated with the outcome of a probability experiment. Random variables are classified as either discrete or continuous based on the nature of their possible values.

• Discrete Random Variable: Takes on a countable set of values, often whole numbers. Typically associated with counting events (e.g., number of successes).

• Continuous Random Variable: Can take on any value within an interval, typically associated with measurements (e.g., weight, time, volume).

Probability Distribution: Describes how probabilities are distributed over the values of the random variable.

• For discrete random variables, probability distributions are often represented by tables or charts.

• For continuous random variables, probability distributions are represented by smooth curves called Probability Density Functions (PDF), where the total area under the curve equals 1.

Common Notation: Random variables are often denoted by capital letters such as X, Y, Z.

Key Properties of Discrete Random Variables

• Mean (Expected Value): The average value expected from the random variable. $\mu = E(X) = \sum x_i p(x_i)$

• Variance: Measures the spread of the random variable's values. $\sigma^2 = \sum (x_i - \mu)^2 p(x_i)$

Key Discrete Random Variables

• Binomial Random Variable:

  • Fixed number of independent trials.

  • Each trial has a constant probability of success (p).

  • Two possible outcomes: success or failure.

  • Counts the number of successes in n trials.

  • Calculator commands: BINOMPDF (probability of exactly X successes), BINOMCDF (probability of X or fewer successes).

• Poisson Random Variable:

  • Counts the number of events in a fixed unit of time or space.

  • Events occur independently and at a constant average rate (lambda).

  • Calculator commands: POISSONPDF (probability of exactly X events), POISSONCDF (probability of X or fewer events).

• Hypergeometric Random Variable:

  • Used when sampling without replacement from a finite population with known composition.

  • Example: Drawing marbles from a jar with known color distribution.

Continuous Random Variables

• Result from measurements; can take any value within an interval.

• Probability distributions are described by Probability Density Functions (PDFs).

• The area under the PDF curve represents probability.

Normal Random Variable

• Bell-shaped probability density function, centered at the population mean.

• Standard deviation marks points of inflection.

• Calculator commands: NormalCDF (probabilities), INVNORM (find values for given percentiles).

Standard Normal Random Variable

• Designated as 'Z', with mean 0 and standard deviation 1.

• Z-scores indicate the number of standard deviations from the mean.

• Useful for comparing different normal distributions (e.g., SAT vs. ACT scores).

Sampling Distributions

Definition and Importance

Sampling distributions describe the probability distribution of a statistic (such as sample mean or sample proportion) computed from a random sample. They are essential for making inferences about population parameters.

• Sample averages (\bar{x}) and sample proportions (\hat{p}) are continuous random variables.

• Each has a probability density function, called a sampling distribution.

Key Properties

• Mean of Sampling Distribution of Sample Mean: $E(\bar{x}) = \mu$

• Standard Deviation of Sampling Distribution of Sample Mean: $\sigma_{\bar{x}} = \frac{\sigma}{\sqrt{n}}$

• Mean of Sampling Distribution of Sample Proportion: $E(\hat{p}) = p$

• Standard Deviation of Sampling Distribution of Sample Proportion: $\sigma_{\hat{p}} = \sqrt{\frac{p(1-p)}{n}}$

Central Limit Theorem (CLT)

• If sample size n ≥ 30, the sampling distribution of the sample mean is approximately normal, regardless of the population's distribution.

• If the population is normal, the sampling distribution of the sample mean is normal for any sample size.

• Larger samples yield tighter sampling distributions (smaller standard error).

Unbiased Estimators and Minimum Variance

• An estimator is unbiased if its expected value equals the population parameter.

• The sample mean (\bar{x}) is an unbiased estimator for the population mean (\mu).

• The sample proportion (\hat{p}) is an unbiased estimator for the population proportion (p).

• An estimator has minimum variance if it has the smallest variance among all unbiased estimators.

Confidence Intervals for Mean (\mu) and Proportion (p)

Definition and Construction

A confidence interval (CI) is a range of values, derived from sample statistics, that is likely to contain the population parameter. The confidence level (e.g., 95%) indicates the probability that the interval contains the parameter in repeated sampling.

• Margin of Error: The half-width of the confidence interval; reflects sampling error.

Large Sample Confidence Interval for Mean (\mu)

• Relies on the Central Limit Theorem; uses sample standard deviation (s) as an estimate for population standard deviation (\sigma).

• Calculator command: Zinterval.

• Critical value (Z): Area in each tail = $\alpha/2$; use INVNORM to find Z.

• Formula: $\bar{x} \pm Z_{\alpha/2} \frac{s}{\sqrt{n}}$

Small Sample Confidence Interval for Mean (\mu)

• Requires evidence that the population is approximately normal.

• Uses t-distribution; calculator command: tinterval.

• Degrees of freedom (DF): $n - 1$.

• Critical value (t): Area in each tail = $\alpha/2$; use INVT with DF.

• Formula: $\bar{x} \pm t_{\alpha/2, df} \frac{s}{\sqrt{n}}$

T Random Variable and Degrees of Freedom

• The t-distribution is a family of curves resembling the standard normal curve.

• Each curve is defined by its degrees of freedom (DF = n - 1).

• As sample size increases, the t-distribution approaches the normal distribution.

Large Sample Confidence Interval for Proportion (p)

• Requires at least 15 successes and 15 failures in the sample.

• Calculator command: 1prop-ZINT.

• Formula: $\hat{p} \pm Z_{\alpha/2} \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}$

Small Sample Confidence Interval for Proportion (p)

• Used when there are fewer than 15 successes or failures; no standard calculator command.

Sample Size Determination

• To achieve a desired margin of error, use $p = 0.5$ if no prior estimate is available.

• Formula provided on formula sheet (not specified here).

Alpha and Confidence Level

• $\alpha = 1 - \text{confidence level}$

• Z and t scores are associated with an area of $\alpha/2$ in each tail.

Hypothesis Testing for Mean (\mu) and Proportion (p)

Key Concepts and Steps

• Null Hypothesis (H0): Represents the status quo; must contain an equal sign.

• Alternative Hypothesis (Ha): Represents the claim to be tested; uses >, <, or ≠.

• Type I Error: Rejecting a true null hypothesis.

• Type II Error: Failing to reject a false null hypothesis.

• Significance Level (Alpha): Probability of Type I error; controls the risk.

• Test Statistic: Converts sample statistic to a Z or t value.

• Critical Value: Z or t score marking the start of the rejection region.

• Rejection Region: Area(s) in the tails of the distribution where H0 is rejected.

• P-value: Probability of observing the sample result (or more extreme) if H0 is true; quantifies how unusual the result is.

Types of Tests

• One-Tailed Test (Left): Ha uses <.

• One-Tailed Test (Right): Ha uses >.

• Two-Tailed Test: Ha uses ≠.

Decision Rules

• If p-value < alpha: Reject H0 and accept Ha.

• If p-value > alpha: Insufficient evidence to reject H0; H0 is plausible.

• If test statistic falls in rejection region: Reject H0 and accept Ha.

Reporting Decisions

• "There is sufficient evidence at an alpha level of xx to reject the null hypothesis and accept the alternative hypothesis."

• "There is insufficient evidence at an alpha level of xx to reject the null hypothesis. Therefore, the null hypothesis is plausible."

Calculator Commands

• ZTEST: Large sample hypothesis test for mean (\mu).

• TTEST: Small sample hypothesis test for mean (\mu), if population is normal.

• 1propZtest: Large sample hypothesis test for proportion (p), requires at least 15 successes and 15 failures.

Summary Table: Hypothesis Test Types

Test Type	Sample Size	Distribution Used	Calculator Command
Mean (\mu), Large Sample	n ≥ 30	Normal	ZTEST
Mean (\mu), Small Sample	n < 30	t-distribution	TTEST
Proportion (p), Large Sample	At least 15 successes and 15 failures	Normal	1propZtest

Example: Hypothesis Test for Mean

• Suppose a manufacturer claims the mean lifetime of a battery is 100 hours. A sample of 40 batteries yields a mean of 98 hours and a standard deviation of 5 hours. Test the claim at alpha = 0.05.

• Null hypothesis: $H_0: \mu = 100$

• Alternative hypothesis: $H_a: \mu \neq 100$

• Test statistic: $z = \frac{\bar{x} - \mu_0}{s/\sqrt{n}}$

• Compare z to critical value or p-value to alpha to make decision.

Additional info: Academic context and formulas have been expanded for clarity and completeness.