Single Sample & One- and Two-Sample Estimation Problems

Single Sample & One- and Two-Sample Estimation Problems

In statistics, estimation involves using sample data to estimate population parameters. Estimation problems can be classified into single-sample problems (where you have data from one sample) and two-sample problems (where you compare two independent samples). These problems are fundamental in hypothesis testing, confidence intervals, and regression analysis.

1. Single-Sample Estimation Problems

A single-sample estimation problem arises when you are given data from a single sample and are tasked with estimating a population parameter, such as the mean or proportion, or making inferences about the population from which the sample was drawn.

a. Estimating the Population Mean ($\mu$)

When the population mean $\mu$ is unknown and we have a sample of data, we can estimate it using the sample mean $\bar{x}$. If the population variance $\sigma^2$ is known, we use the normal distribution; if unknown, we use the t-distribution.

1. Confidence Interval for the Mean (When $\sigma$ is known):

   • Formula: $$\mu = \bar{x} \pm z_{\alpha/2} \times \frac{\sigma}{\sqrt{n}}$$ Where:
     • $\bar{x}$ = sample mean,
     • $\sigma$ = population standard deviation (known),
     • $n$ = sample size,
     • $z_{\alpha/2}$ = z-value for the confidence level (e.g., for a 95% confidence interval, $z = 1.96$).

2. Confidence Interval for the Mean (When $\sigma$ is unknown):

   • Formula: $$\mu = \bar{x} \pm t_{\alpha/2} \times \frac{s}{\sqrt{n}}$$ Where:
     • $\bar{x}$ = sample mean,
     • $s$ = sample standard deviation,
     • $n$ = sample size,
     • $t_{\alpha/2}$ = t-value for the confidence level with $n-1$ degrees of freedom.

b. Estimating the Population Proportion ($p$)

When estimating the proportion of a population that has a certain characteristic, the sample proportion $\hat{p}$ is used to estimate the population proportion $p$.

• Confidence Interval for the Proportion:
  • Formula: $$p = \hat{p} \pm z_{\alpha/2} \times \sqrt{\frac{\hat{p}(1 - \hat{p})}{n}}$$ Where:
    • $\hat{p}$ = sample proportion,
    • $n$ = sample size,
    • $z_{\alpha/2}$ = z-value for the desired confidence level (e.g., for a 95% confidence interval, $z = 1.96$).

2. One-Sample Hypothesis Testing

One-sample hypothesis testing involves using sample data to test hypotheses about a population parameter (such as the population mean or proportion). The two main types of tests are one-sample z-tests and one-sample t-tests.

a. One-Sample z-Test for the Mean (When Population Variance is Known)

• Hypothesis:

  • Null hypothesis ($H_0$): $\mu = \mu_0$ (Population mean equals a specific value)
  • Alternative hypothesis ($H_A$): $\mu \neq \mu_0$ (Population mean is different from the specific value)

• Test Statistic:

  $$z = \frac{\bar{x} - \mu_0}{\frac{\sigma}{\sqrt{n}}}$$

  Where:

  • $\bar{x}$ = sample mean,
  • $\mu_0$ = hypothesized population mean,
  • $\sigma$ = known population standard deviation,
  • $n$ = sample size.

• Decision Rule:

  • If $|z| > z_{\alpha/2}$, reject $H_0$.

b. One-Sample t-Test for the Mean (When Population Variance is Unknown)

When the population variance is unknown, we use the t-distribution for hypothesis testing.

• Hypothesis:

  • Null hypothesis ($H_0$): $\mu = \mu_0$
  • Alternative hypothesis ($H_A$): $\mu \neq \mu_0$

• Test Statistic:

  $$t = \frac{\bar{x} - \mu_0}{\frac{s}{\sqrt{n}}}$$

  Where:

  • $\bar{x}$ = sample mean,
  • $\mu_0$ = hypothesized population mean,
  • $s$ = sample standard deviation,
  • $n$ = sample size.

• Decision Rule:

  • If $|t| > t_{\alpha/2, n-1}$, reject $H_0$.

3. Two-Sample Estimation Problems

In two-sample estimation problems, we compare the means or proportions of two independent samples to draw inferences about the population parameters. The two main types are two-sample z-tests for means (when population variances are known) and two-sample t-tests for means (when population variances are unknown), as well as tests for comparing proportions.

a. Two-Sample z-Test for the Difference Between Means (When Population Variances are Known)

• Hypothesis:

  • Null hypothesis ($H_0$): $\mu_1 = \mu_2$ (The two population means are equal)
  • Alternative hypothesis ($H_A$): $\mu_1 \neq \mu_2$

• Test Statistic:

  $$z = \frac{\bar{x}_1 - \bar{x}_2}{\sqrt{\frac{\sigma_1^2}{n_1} + \frac{\sigma_2^2}{n_2}}}$$

  Where:

  • $\bar{x}_1$ and $\bar{x}_2$ are the sample means,
  • $\sigma_1^2$ and $\sigma_2^2$ are the population variances (known),
  • $n_1$ and $n_2$ are the sample sizes.

• Decision Rule:

  • If $|z| > z_{\alpha/2}$, reject $H_0$.

b. Two-Sample t-Test for the Difference Between Means (When Population Variances are Unknown)

• Hypothesis:

  • Null hypothesis ($H_0$): $\mu_1 = \mu_2$
  • Alternative hypothesis ($H_A$): $\mu_1 \neq \mu_2$

• Test Statistic:

  $$t = \frac{\bar{x}_1 - \bar{x}_2}{\sqrt{\frac{s_1^2}{n_1} + \frac{s_2^2}{n_2}}}$$

  Where:

  • $\bar{x}_1$ and $\bar{x}_2$ are the sample means,
  • $s_1^2$ and $s_2^2$ are the sample variances,
  • $n_1$ and $n_2$ are the sample sizes.

• Degrees of Freedom: The degrees of freedom (df) for this test is calculated using the Welch-Satterthwaite equation:

  $$df = \frac{\left( \frac{s_1^2}{n_1} + \frac{s_2^2}{n_2} \right)^2}{\frac{\left( \frac{s_1^2}{n_1} \right)^2}{n_1 - 1} + \frac{\left( \frac{s_2^2}{n_2} \right)^2}{n_2 - 1}}$$

• Decision Rule:

  • If $|t| > t_{\alpha/2, df}$, reject $H_0$.

c. Two-Sample z-Test for Proportions

When comparing two proportions, we use the two-sample z-test for proportions.

• Hypothesis:

  • Null hypothesis ($H_0$): $p_1 = p_2$ (The two population proportions are equal)
  • Alternative hypothesis ($H_A$): $p_1 \neq p_2$

• Test Statistic:

  $$z = \frac{\hat{p}_1 - \hat{p}_2}{\sqrt{\hat{p}(1 - \hat{p}) \left( \frac{1}{n_1} + \frac{1}{n_2} \right)}}$$

  Where:

  • $\hat{p}_1$ and $\hat{p}_2$ are the sample proportions,
  • $\hat{p} = \frac{x_1 + x_2}{n_1 + n_2}$ is the pooled sample proportion,
  • $n_1$ and $n_2$ are the sample sizes.

• Decision Rule:

  • If $|z| > z_{\alpha/2}$, reject $H_0$.

Summary

• Single-sample estimation problems involve estimating population parameters (mean or proportion) from a single sample, using confidence intervals or hypothesis tests.
• One-sample hypothesis tests (e.g., z-tests, t-tests) test if a population parameter (mean or proportion) is equal to a specific value.
• Two-sample estimation problems involve comparing two independent samples to estimate and compare their population parameters (means or proportions). Common tests include the two-sample z-test and t-test for means and the two-sample z-test for proportions.