Are there any practical cases or a general rule about this?

**Contents**hide

#### Best Answer

As an aside, this depends on how you define a "good estimator".

Most simply, the sample median is a good estimator of the population mean when the population mean and population median are equal. If the population mean and population median are different, then the sample median estimates the population median and will likely not do a good job of estimating the population mean.

For symmetric distributions, the population mean and population median are equal, so the sample median can be used to estimate the population mean. This will yield a more robust estimator compared to the sample mean.

### Similar Posts:

- Solved – Expectation of the sample median for symmetric distributions
- Solved – Variance of sample median
- Solved – Sample mean is always an optimal estimator of the mean
- Solved – How to interpret Standard deviation or 95% CI to describe the population
- Solved – Why can we assume a single sample is normally distributed