What is the fundamental definition of an unbiased estimator?

An estimator is unbiased if the mean of its sampling distribution is equal to the population parameter it is estimating. This implies that, on average, the estimator hits the true target value without systematic deviation.

How does increasing the sample size $n$ affect the bias of the sample mean?

Increasing the sample size does not affect the bias; the sample mean remains an unbiased estimator regardless of $n$. However, a larger sample size reduces the variability, making the estimates more precise and closer to the true mean.

What is the difference between an unbiased estimator and a biased one in terms of their sampling distributions?

The sampling distribution of an unbiased estimator is centered exactly at the population parameter. In contrast, the sampling distribution of a biased estimator is centered at a value either higher or lower than the true parameter.

Why are the sample median and sample mode generally considered biased estimators for the population mean in skewed distributions?

In skewed distributions, the median and mode do not align with the mean. Therefore, the average value of all possible sample medians or modes will not equal the true population mean, resulting in systematic error.

What is a common error when interpreting the 'unbiased' nature of a single sample mean?

A common error is believing that a single sample mean must be equal to the population mean. Unbiasedness only means the average of *all* possible sample means equals the population mean; any single sample mean will likely differ due to sampling variability.

If an estimator is unbiased, does it mean it is always the best choice for a study?

Not necessarily. While it is accurate on average, an unbiased estimator with very high variability might be less useful than a slightly biased estimator with very low variability, as the latter might provide more consistent results.

What happens to the standard deviation of an unbiased estimator as the sample size increases?

The standard deviation of the estimator (the standard error) decreases as the sample size increases. This 'squashes' the sampling distribution, making it taller and narrower around the true population parameter.

Which three specific sample statistics are typically identified as unbiased estimators of their corresponding population parameters?

The sample mean ($\bar{x}$), the sample proportion ($\hat{p}$), and the sample standard deviation ($s$) are the three primary unbiased estimators used to predict $\mu$, $p$, and $\sigma$ respectively.

What error occurs if a student claims that 'bias decreases as sample size increases'?

This is a misconception because bias is a property of the estimator's center relative to the parameter. If an estimator is biased, it remains biased regardless of sample size; only the variability (spread) of the estimates decreases with a larger sample.

How can you identify an unbiased estimator from a set of different sampling distribution graphs?

Look for the distribution whose center (mean) is located exactly at the value of the population parameter. If the center of the distribution is shifted to the left or right of the parameter, the estimator is biased.

Library Podcasts

Courses

Referral & Rewards

Revision Notes

College Board

Statistics

Unit 5: Sampling Distributions

Biased & Unbiased Estimators

Summary

In statistical inference, estimators are sample statistics used to predict unknown population parameters. An estimator is considered unbiased if the mean of its sampling distribution perfectly aligns with the true population parameter, ensuring that the estimates are centered correctly on average.

1. Definition & Core Concepts

Estimator: A specific type of sample statistic, such as a mean or proportion, that is used to provide a numerical guess for a population parameter.
Estimate: The actual numerical value resulting from applying an estimator to a specific sample of data.
Unbiased Estimator: A statistic whose sampling distribution has a mean equal to the value of the population parameter being estimated.
Biased Estimator: A statistic that consistently overestimates or underestimates the population parameter because the mean of its sampling distribution does not equal the true parameter value.

A comparison of two sampling distributions: one centered exactly on the true population parameter (unbiased) and one shifted to the right (biased).

2. Underlying Principles

3. Key Unbiased Estimators

4. Impact of Sample Size

Mean Stability: Changing the sample size $n$ does not change the mean of the sampling distribution for an unbiased estimator. It remains equal to the population parameter.
Variability Reduction: Increasing the sample size $n$ reduces the standard deviation (standard error) of the estimator. This means that while the estimator remains unbiased, the estimates become more tightly clustered around the true parameter.
Efficiency: The 'best' unbiased estimators are those that have the smallest possible standard deviation for a given sample size, as they provide more reliable predictions.

5. Key Distinctions

6. Exam Strategy & Tips