本页对应的英文页面已更新,但尚未翻译。 若要查看最新内容,请点击此处访问英文页面。

正态分布

拟合和计算正态(高斯)分布,生成该分布的随机样本

Statistics and Machine Learning Toolbox™ 提供了几种处理正态分布的方法。

  • 可通过对样本数据进行概率分布拟合或通过指定参数值来创建概率分布对象 NormalDistribution。然后使用对象函数来计算分布、生成随机数等。

  • 使用 Distribution Fitter App 以交互方式处理正态分布。您可以从该 App 中导出对象并使用对象函数。

  • 将分布特定的函数与指定的分布参数结合使用。分布特定的函数可以接受多个正态分布的参数。

  • 将一般分布函数(cdficdfpdfrandom)与指定的分布名称 ('Normal') 和参数结合使用。

要了解正态分布,请参阅Normal Distribution

对象

NormalDistributionNormal probability distribution object

App

Distribution FitterFit probability distributions to data

函数

全部展开

创建 NormalDistribution 对象

makedistCreate probability distribution object
fitdistFit probability distribution object to data

使用 NormalDistribution 对象

cdfCumulative distribution function
icdfInverse cumulative distribution function
iqrInterquartile range
meanMean of probability distribution
medianMedian of probability distribution
negloglikNegative loglikelihood of probability distribution
paramciConfidence intervals for probability distribution parameters
pdfProbability density function
proflikProfile likelihood function for probability distribution
randomRandom numbers
stdStandard deviation of probability distribution
truncateTruncate probability distribution object
varVariance of probability distribution
normcdfNormal cumulative distribution function
normpdfNormal probability density function
norminvNormal inverse cumulative distribution function
normlikeNormal negative loglikelihood
normstatNormal mean and variance
normfitNormal parameter estimates
normrndNormal random numbers
mleMaximum likelihood estimates
mlecovAsymptotic covariance of maximum likelihood estimators
histfitHistogram with a distribution fit
normplotNormal probability plot
normspecNormal density plot shading between specifications
概率分布函数Interactive density and distribution plots
qqplotQuantile-quantile plot
randtoolInteractive random number generation

主题

Normal Distribution

Learn about the normal distribution. The normal distribution is a two-parameter (mean and standard deviation) family of curves. Central Limit Theorem states that the normal distribution models the sum of independent samples from any distribution as the sample size goes to infinity.