Point estimation and interval estimation

Point estimation

If you use 100 The average height of middle school students in China is 150cm, That's the point estimate ,150cm It's a point estimator . Point estimation is to estimate a value of a parameter , Interval estimation is an interval in which parameters are estimated .

interval estimation

If you don't want to use an average of samples to estimate the overall average （ such as 150cm）, But use an interval to estimate （ such as 140-155cm）, This is called interval estimation . Interval estimation leaves more fault-tolerant space than point estimation .
The range is large , It can be predicted that the height is 149-151cm Between , It can also be predicted that 140-160cm Between , It could be something else . But you'll see , The former has a lower probability of accurate prediction than the latter , Because the range of its prediction is too narrow ; The latter has a higher probability of accurate prediction , Because its prediction range is wider .
It's also like rolling dice , If your prediction is 3-6, Xiaoming prediction 3-4, Then the probability that you guessed right is 67%, The probability that Xiao Ming guesses right is 33%, You are more likely to guess right than Xiao Ming .

Confidence interval and confidence level

Confidence level ： The probability that the parameter falls on the confidence interval .
confidence interval ： Is an estimation interval of the parameter , The higher the confidence level , The wider the confidence interval , The lower the confidence level , The narrower the confidence interval .

Example