How to read bootstrap test

The Bootstrap test estimates the sampling distribution and evaluates its statistical significance by repeatedly sampling and calculating statistics. The steps include: randomly sampling from the original data with replacement. Calculate statistics and repeat multiple times. Create bootstrapped samples and sampling distributions of statistics. Calculate the P value, which measures the probability of falling on the observed statistic or a more extreme value. The smaller the P value, the higher the statistical significance: P value

Bootstrap test

The Bootstrap test is a statistical method used to evaluate the sampling distribution of a statistic to determine whether it is statistically significant . The following are the steps of the Bootstrap test:

1. Extract a sample from the original data set: Randomly select a sample of the size of the original data set from the original data set with replacement. That is, the extracted elements can appear repeatedly in the sample.
2. Calculate statistics: Calculate the statistics of interest, such as mean, median or standard deviation, on the extracted samples.
3. Repeat steps 1 and 2: Repeat steps 1 and 2 multiple times to create many samples and calculate the corresponding statistics. These samples are called bootstrapped samples, and the statistics calculated are called bootstrapped statistics.
4. Create sampling distribution: Collect the bootstrapped statistics to create a sampling distribution. The sampling distribution shows how a statistic will change if you repeat sampling and calculating the statistic many times.
5. Calculate P-value: The P-value is the probability of falling on the observed statistic or a more extreme statistic. The smaller the P value, the greater the suspicion that the observed statistic was produced by random sampling.

P value explanation

P value is often used as a measure of statistical significance. Based on commonly accepted thresholds, P-values:

• ##P-values ​​ are considered statistically significant, indicating that the observed statistic is unlikely to result from random sampling.
• 0.05 <= P value < 0.1: Considered close to significance, but statistical significance cannot be clearly determined.
• P value >= 0.1: is considered not significant, indicating that the observed statistics may be generated by random sampling.

It should be noted that the Bootstrap test is a sampling method, and its results depend on the Bootstrapped sample. Therefore, the Bootstrap test may not always be completely accurate, but it can usually provide a good estimate of the sampling distribution of the statistic.

The above is the detailed content of How to read bootstrap test. For more information, please follow other related articles on the PHP Chinese website!