Fundamentals of Hypothesis Testing: One-Sample Tests (Business Calculus Context)

Hypothesis Testing: One-Sample Tests

Introduction to Hypothesis Testing

Hypothesis testing is a fundamental statistical method used to make inferences about population parameters based on sample data. In business calculus, it is essential for decision-making and evaluating claims about means or proportions.

• Hypothesis: A claim or assertion about a population parameter, such as the population mean ($\mu$) or population proportion ($\pi$).

• Example: The mean monthly cell phone bill in a city is $\mu = $42.

• Example: The proportion of adults in a city with cell phones is $\pi = 0.88$.

The Null Hypothesis ($H_0$)

The null hypothesis is the default assumption or claim to be tested. It is always about a population parameter, not a sample statistic.

• Form: $H_0: \mu = \text{value}$ or $H_0: \pi = \text{value}$

• Example: The mean diameter of a manufactured bolt is 30mm ($H_0: \mu = 30$).

• Assume $H_0$ is true at the start, similar to 'innocent until proven guilty.'

• Represents the current belief or status quo.

• Always contains an equality sign ($=$, $\leq$, or $\geq$).

The Alternative Hypothesis ($H_1$)

The alternative hypothesis is the claim that contradicts the null hypothesis. It is what the researcher typically aims to support.

• Form: $H_1: \mu \neq \text{value}$, $H_1: \mu > \text{value}$, or $H_1: \mu < \text{value}$

• Example: The mean diameter of a manufactured bolt is not equal to 30mm ($H_1: \mu \neq 30$).

• Challenges the status quo.

• Never contains an equality sign.

The Hypothesis Testing Process

Hypothesis testing involves comparing sample statistics to population parameters to determine if there is enough evidence to reject the null hypothesis.

• State the claim (e.g., population mean age is 50: $H_0: \mu = 50$).

• Sample the population and calculate the sample mean ($\bar{X}$).

• If the sample mean is significantly different from the claimed mean, the null hypothesis may be rejected.

• Use probability to assess how likely the observed sample mean is, given $H_0$ is true.

Test Statistic and Critical Values

The test statistic quantifies the difference between the sample statistic and the population parameter. Critical values define the threshold for rejecting $H_0$.

• If the sample mean is close to the population mean, do not reject $H_0$.

• If the sample mean is far from the population mean, reject $H_0$.

• Critical values create a 'line in the sand' for decision-making.

Risks in Decision Making: Type I and Type II Errors

Errors can occur in hypothesis testing, and understanding their probabilities is crucial for sound decision-making.

• Type I Error ($\alpha$): Rejecting a true null hypothesis (false alarm). Probability is $\alpha$, the level of significance.

• Type II Error ($\beta$): Failing to reject a false null hypothesis (missed opportunity). Probability is $\beta$.

Possible Hypothesis Test Outcomes

Decision	$H_0$ True	$H_0$ False
Do Not Reject $H_0$	Correct Decision Confidence = $1 - \alpha$	Type II Error $P$(Type II Error) = $\beta$
Reject $H_0$	Type I Error $P$(Type I Error) = $\alpha$	Correct Decision Power = $1 - \beta$

• Confidence coefficient ($1-\alpha$): Probability of not rejecting $H_0$ when it is true.

• Confidence level: $(1-\alpha) \times 100\%$

• Power of a test ($1-\beta$): Probability of rejecting $H_0$ when it is false.

Type I & II Error Relationship

• Type I and Type II errors cannot occur simultaneously.

• Increasing $\alpha$ (Type I error probability) decreases $\beta$ (Type II error probability), and vice versa.

Factors Affecting Type II Error ($\beta$)

• $\beta$ decreases as the difference between the hypothesized parameter and its true value increases.

• $\beta$ decreases as $\alpha$ increases.

• $\beta$ decreases as sample size ($n$) increases.

• $\beta$ increases as population standard deviation ($\sigma$) increases.

Level of Significance and the Rejection Region

The level of significance ($\alpha$) determines the size of the rejection region in hypothesis testing.

• For a two-tail test: $H_0: \mu = 30$, $H_1: \mu \neq 30$

• Rejection regions are in both tails of the sampling distribution, each with area $\alpha/2$.

Hypothesis Tests for the Mean

There are two main types of tests for the population mean, depending on whether the population standard deviation ($\sigma$) is known or unknown.

• $\sigma$ Known: Use the Z test.

• $\sigma$ Unknown: Use the t test.

Z Test of Hypothesis for the Mean ($\sigma$ Known)

When the population standard deviation is known, the Z test is used to test hypotheses about the mean.

• Convert the sample mean ($\bar{X}$) to a Z test statistic:

$Z_{STAT} = \frac{\bar{X} - \mu}{\sigma / \sqrt{n}}$

Critical Value Approach to Testing

This approach uses critical values to determine whether to reject the null hypothesis.

• For a two-tail test, determine the critical Z values for the specified $\alpha$.

• If $Z_{STAT}$ falls in the rejection region, reject $H_0$; otherwise, do not reject $H_0$.

Steps in the Critical Value Approach

1. State the null and alternative hypotheses.

2. Choose the level of significance ($\alpha$) and sample size ($n$).

3. Determine the appropriate test statistic and sampling distribution.

4. Determine the critical values dividing rejection and nonrejection regions.

5. Collect sample data and compute the test statistic.

6. Make the statistical decision and interpret the result in context.

Example: Hypothesis Testing for the Mean

• Claim: The true mean diameter of a manufactured bolt is 30mm ($\sigma = 0.8$).

• Hypotheses: $H_0: \mu = 30$, $H_1: \mu \neq 30$ (two-tail test).

• Significance level: $\alpha = 0.05$, $n = 100$.

• Critical Z values: $\pm 1.96$.

• Sample results: $\bar{X} = 29.84$, $\sigma = 0.8$.

• Test statistic:

$Z_{STAT} = \frac{29.84 - 30}{0.8 / \sqrt{100}} = \frac{-0.16}{0.08} = -2.0$

• Since $Z_{STAT} = -2.0 < -1.96$, reject $H_0$ and conclude there is sufficient evidence that the mean diameter is not equal to 30mm.

p-Value Approach to Testing

The p-value is the probability of obtaining a test statistic as extreme as the observed value, assuming $H_0$ is true.

• If p-value $< \alpha$, reject $H_0$.

• If p-value $\geq \alpha$, do not reject $H_0$.

Steps in the p-Value Approach

1. State the null and alternative hypotheses.

2. Choose the level of significance ($\alpha$) and sample size ($n$).

3. Determine the appropriate test statistic and sampling distribution.

4. Collect sample data, compute the test statistic and p-value.

5. Make the statistical decision and interpret the result in context.

Example: p-Value Hypothesis Testing

• Claim: The true mean diameter of a manufactured bolt is 30mm ($\sigma = 0.8$).

• Hypotheses: $H_0: \mu = 30$, $H_1: \mu \neq 30$ (two-tail test).

• Significance level: $\alpha = 0.05$, $n = 100$.

• Sample results: $\bar{X} = 29.84$, $\sigma = 0.8$.

• Test statistic: $Z_{STAT} = -2.0$

• p-value calculation: For $Z = -2.0$, $P(Z < -2.0) = 0.0228$; for two tails, p-value = $0.0228 + 0.0228 = 0.0456$.

• Since p-value $= 0.0456 < 0.05$, reject $H_0$ and conclude there is sufficient evidence that the mean diameter is not equal to 30mm.

Additional info: These notes are based on business statistics slides but are directly relevant to Business Calculus students, especially in applications of calculus to probability and statistical inference.