九.Stats模块
1.概率分布(Distribution)
(1)基类(Base Class):
连续型随机变量的基类:class scipy.stats.rv_continuous([momtype=1,a=None,b=None,xtol=1e-14,badvalue=None,name=None,longname=None,shapes=None,extradoc=None,seed=None])
######################################################################################################################
离散型随机变量的基类:class scipy.stats.rv_discrete([a=0,b=inf,name=None,badvalue=None,moment_tol=1e-08,values=None,inc=1,longname=None,shapes=None,extradoc=None,seed=None])
######################################################################################################################
直方图给出的分布:class scipy.stats.rv_histogram(<histogram>[,*args,**kwargs])
(2)连续型随机变量(Continuous Distribution):
scipy.stats.alpha(*args, **kwds):An alpha continuous random variable.
scipy.stats.anglit(*args, **kwds):An anglit continuous random variable.
scipy.stats.arcsine(*args, **kwds):An arcsine continuous random variable.
scipy.stats.argus(*args, **kwds):Argus distribution
scipy.stats.beta(*args, **kwds):A beta continuous random variable.
scipy.stats.betaprime(*args, **kwds):A beta prime continuous random variable.
scipy.stats.bradford(*args, **kwds):A Bradford continuous random variable.
scipy.stats.burr(*args, **kwds):A Burr (Type III) continuous random variable.
scipy.stats.burr12(*args, **kwds):A Burr (Type XII) continuous random variable.
scipy.stats.cauchy(*args, **kwds):A Cauchy continuous random variable.
scipy.stats.chi(*args, **kwds):A chi continuous random variable.
scipy.stats.chi2(*args, **kwds):A chi-squared continuous random variable.
scipy.stats.cosine(*args, **kwds):A cosine continuous random variable.
scipy.stats.crystalball(*args, **kwds):Crystalball distribution
scipy.stats.dgamma(*args, **kwds):A double gamma continuous random variable.
scipy.stats.dweibull(*args, **kwds):A double Weibull continuous random variable.
scipy.stats.erlang(*args, **kwds):An Erlang continuous random variable.
scipy.stats.expon(*args, **kwds):An exponential continuous random variable.
scipy.stats.exponnorm(*args, **kwds):An exponentially modified Normal continuous random variable.
scipy.stats.exponweib(*args, **kwds):An exponentiated Weibull continuous random variable.
scipy.stats.exponpow(*args, **kwds):An exponential power continuous random variable.
scipy.stats.f(*args, **kwds):An F continuous random variable.
scipy.stats.fatiguelife(*args, **kwds):A fatigue-life (Birnbaum-Saunders) continuous random variable.
scipy.stats.fisk(*args, **kwds):A Fisk continuous random variable.
scipy.stats.foldcauchy(*args, **kwds):A folded Cauchy continuous random variable.
scipy.stats.foldnorm(*args, **kwds):A folded normal continuous random variable.
scipy.stats.genlogistic(*args, **kwds):A generalized logistic continuous random variable.
scipy.stats.gennorm(*args, **kwds):A generalized normal continuous random variable.
scipy.stats.genpareto(*args, **kwds):A generalized Pareto continuous random variable.
scipy.stats.genexpon(*args, **kwds):A generalized exponential continuous random variable.
scipy.stats.genextreme(*args, **kwds):A generalized extreme value continuous random variable.
scipy.stats.gausshyper(*args, **kwds):A Gauss hypergeometric continuous random variable.
scipy.stats.gamma(*args, **kwds):A gamma continuous random variable.
scipy.stats.gengamma(*args, **kwds):A generalized gamma continuous random variable.
scipy.stats.genhalflogistic(*args, **kwds):A generalized half-logistic continuous random variable.
scipy.stats.geninvgauss(*args, **kwds):A Generalized Inverse Gaussian continuous random variable.
scipy.stats.gilbrat(*args, **kwds):A Gilbrat continuous random variable.
scipy.stats.gompertz(*args, **kwds):A Gompertz (or truncated Gumbel) continuous random variable.
scipy.stats.gumbel_r(*args, **kwds):A right-skewed Gumbel continuous random variable.
scipy.stats.gumbel_l(*args, **kwds):A left-skewed Gumbel continuous random variable.
scipy.stats.halfcauchy(*args, **kwds):A Half-Cauchy continuous random variable.
scipy.stats.halflogistic(*args, **kwds):A half-logistic continuous random variable.
scipy.stats.halfnorm(*args, **kwds):A half-normal continuous random variable.
scipy.stats.halfgennorm(*args, **kwds):The upper half of a generalized normal continuous random variable.
scipy.stats.hypsecant(*args, **kwds):A hyperbolic secant continuous random variable.
scipy.stats.invgamma(*args, **kwds):An inverted gamma continuous random variable.
scipy.stats.invgauss(*args, **kwds):An inverse Gaussian continuous random variable.
scipy.stats.invweibull(*args, **kwds):An inverted Weibull continuous random variable.
scipy.stats.johnsonsb(*args, **kwds):A Johnson SB continuous random variable.
scipy.stats.johnsonsu(*args, **kwds):A Johnson SU continuous random variable.
scipy.stats.kappa4(*args, **kwds):Kappa 4 parameter distribution.
scipy.stats.kappa3(*args, **kwds):Kappa 3 parameter distribution.
scipy.stats.ksone(*args, **kwds):Kolmogorov-Smirnov one-sided test statistic distribution.
scipy.stats.kstwo(*args, **kwds):Kolmogorov-Smirnov two-sided test statistic distribution.
scipy.stats.kstwobign(*args, **kwds):Limiting distribution of scaled Kolmogorov-Smirnov two-sided test statistic.
scipy.stats.laplace(*args, **kwds):A Laplace continuous random variable.
scipy.stats.laplace_asymmetric(*args, **kwds):An asymmetric Laplace continuous random variable.
scipy.stats.levy(*args, **kwds):A Levy continuous random variable.
scipy.stats.levy_l(*args, **kwds):A left-skewed Levy continuous random variable.
scipy.stats.levy_stable(*args, **kwds):A Levy-stable continuous random variable.
scipy.stats.logistic(*args, **kwds):A logistic (or Sech-squared) continuous random variable.
scipy.stats.loggamma(*args, **kwds):A log gamma continuous random variable.
scipy.stats.loglaplace(*args, **kwds):A log-Laplace continuous random variable.
scipy.stats.lognorm(*args, **kwds):A lognormal continuous random variable.
scipy.stats.loguniform(*args, **kwds):A loguniform or reciprocal continuous random variable.
scipy.stats.lomax(*args, **kwds):A Lomax (Pareto of the second kind) continuous random variable.
scipy.stats.maxwell(*args, **kwds):A Maxwell continuous random variable.
scipy.stats.mielke(*args, **kwds):A Mielke Beta-Kappa / Dagum continuous random variable.
scipy.stats.moyal(*args, **kwds):A Moyal continuous random variable.
scipy.stats.nakagami(*args, **kwds):A Nakagami continuous random variable.
scipy.stats.ncx2(*args, **kwds):A non-central chi-squared continuous random variable.
scipy.stats.ncf(*args, **kwds):A non-central F distribution continuous random variable.
scipy.stats.nct(*args, **kwds):A non-central Student’s t continuous random variable.
scipy.stats.norm(*args, **kwds):A normal continuous random variable.
scipy.stats.norminvgauss(*args, **kwds):A Normal Inverse Gaussian continuous random variable.
scipy.stats.pareto(*args, **kwds):A Pareto continuous random variable.
scipy.stats.pearson3(*args, **kwds):A pearson type III continuous random variable.
scipy.stats.powerlaw(*args, **kwds):A power-function continuous random variable.
scipy.stats.powerlognorm(*args, **kwds):A power log-normal continuous random variable.
scipy.stats.powernorm(*args, **kwds):A power normal continuous random variable.
scipy.stats.rdist(*args, **kwds):An R-distributed (symmetric beta) continuous random variable.
scipy.stats.rayleigh(*args, **kwds):A Rayleigh continuous random variable.
scipy.stats.rice(*args, **kwds):A Rice continuous random variable.
scipy.stats.recipinvgauss(*args, **kwds):A reciprocal inverse Gaussian continuous random variable.
scipy.stats.semicircular(*args, **kwds):A semicircular continuous random variable.
scipy.stats.skewnorm(*args, **kwds):A skew-normal random variable.
scipy.stats.t(*args, **kwds):A Student’s t continuous random variable.
scipy.stats.trapezoid(*args, **kwds):A trapezoidal continuous random variable.
scipy.stats.triang(*args, **kwds):A triangular continuous random variable.
scipy.stats.truncexpon(*args, **kwds):A truncated exponential continuous random variable.
scipy.stats.truncnorm(*args, **kwds):A truncated normal continuous random variable.
scipy.stats.tukeylambda(*args, **kwds):A Tukey-Lamdba continuous random variable.
scipy.stats.uniform(*args, **kwds):A uniform continuous random variable.
scipy.stats.vonmises(*args, **kwds):A Von Mises continuous random variable.
scipy.stats.vonmises_line(*args, **kwds):A Von Mises continuous random variable.
scipy.stats.wald(*args, **kwds):A Wald continuous random variable.
scipy.stats.weibull_min(*args, **kwds):Weibull minimum continuous random variable.
scipy.stats.weibull_max(*args, **kwds):Weibull maximum continuous random variable.
scipy.stats.wrapcauchy(*args, **kwds):A wrapped Cauchy continuous random variable.
- 公共方法:
rvs: Random Variates
pdf: Probability Density Function
cdf: Cumulative Distribution Function
sf: Survival Function (1-CDF)
ppf: Percent Point Function (Inverse of CDF)
isf: Inverse Survival Function (Inverse of SF)
stats: Return mean, variance, (Fisher’s) skew, or (Fisher’s) kurtosis
moment: non-central moments of the distribution
(3)离散型随机变量(Discrete Distribution):
scipy.stats.bernoulli(*args, **kwds):A Bernoulli discrete random variable.
scipy.stats.betabinom(*args, **kwds):A beta-binomial discrete random variable.
scipy.stats.binom(*args, **kwds):A binomial discrete random variable.
scipy.stats.boltzmann(*args, **kwds):A Boltzmann (Truncated Discrete Exponential) random variable.
scipy.stats.dlaplace(*args, **kwds):A Laplacian discrete random variable.
scipy.stats.geom(*args, **kwds):A geometric discrete random variable.
scipy.stats.hypergeom(*args, **kwds):A hypergeometric discrete random variable.
scipy.stats.logser(*args, **kwds):A Logarithmic (Log-Series, Series) discrete random variable.
scipy.stats.nbinom(*args, **kwds):A negative binomial discrete random variable.
scipy.stats.nhypergeom(*args, **kwds):A negative hypergeometric discrete random variable.
scipy.stats.planck(*args, **kwds):A Planck discrete exponential random variable.
scipy.stats.poisson(*args, **kwds):A Poisson discrete random variable.
scipy.stats.randint(*args, **kwds):A uniform discrete random variable.
scipy.stats.skellam(*args, **kwds):A Skellam discrete random variable.
scipy.stats.yulesimon(*args, **kwds):A Yule-Simon discrete random variable.
scipy.stats.zipf(*args, **kwds):A Zipf discrete random variable.
(4)多元分布(Continuous Distribution):
scipy.stats.dirichlet(alpha[, seed]):A Dirichlet random variable.
scipy.stats.invwishart([df, scale, seed]):An inverse Wishart random variable.
scipy.stats.matrix_normal([mean, rowcov, colcov, seed]):A matrix normal random variable.
scipy.stats.multinomial(n, p[, seed]):A multinomial random variable.
scipy.stats.multivariate_hypergeom(m, n[, seed]):A multivariate hypergeometric random variable.
scipy.stats.multivariate_normal([mean, cov, allow_singular, seed]):A multivariate normal random variable.
scipy.stats.multivariate_t([loc, shape, df, allow_singular, seed]):A multivariate t-distributed random variable.
scipy.stats.ortho_group:A matrix-valued O(N) random variable.
scipy.stats.random_correlation:A random correlation matrix.
scipy.stats.special_ortho_group([dim, seed]):A matrix-valued SO(N) random variable.
scipy.stats.unitary_group:A matrix-valued U(N) random variable.
scipy.stats.wishart([df, scale, seed]):A Wishart random variable.
2.数据集变换(Transformations):
进行"Box-Cox变换"(Box-Cox Transformation):[<boxcox>,<maxlog>,<interval>=]scipy.stats.boxcox(<x>[,lmbda=None,alpha=None])
求最佳Box-Cox变换参数:[<maxlog>=]scipy.stats.boxcox_normmax(<x>[,brack=(-2.0,2.0),method='pearsonr'])
求Box-Cox对数似然函数值:[<llf>=]scipy.stats.boxcox_llf(<lmb>,<data>)
######################################################################################################################
进行"Yeo-Johnson变换"(Yeo-Johnson Transformation):[<yeojohnson>,<maxlog>=]scipy.stats.yeojohnson(<x>[,lmbda=None])
求最佳Yeo-Johnson变换参数:[<maxlog>=]scipy.stats.yeojohnson_normmax(<x>[,brack=(-2,2)])
求Yeo-Johnson对数似然函数值:[<llf>=]scipy.stats.yeojohnson_llf(<lmb>,<data>)
######################################################################################################################
进行"O'Brien变换"(O'Brien Transform):[<obrientransform>=]scipy.stats.obrientransform([*args])
######################################################################################################################
进行"Iterative Sigma-Clipping":[<clipped>,<lower>,<upper>=]scipy.stats.sigmaclip(<a>[,low=4.0,high=4.0])
######################################################################################################################
从2侧分别裁剪掉指定比例的数据:[<out>=]scipy.stats.trimboth(<a>,<proportiontocut>[,axis=0])
从1侧裁剪掉指定比例的数据:[<trim1>=]scipy.stats.trim1(<a>,<proportiontocut>[,tail='right',axis=0])
######################################################################################################################
求"Z分数"(Z-Score):[<zscore>=]scipy.stats.zscore(<a>[,axis=0,ddof=0,nan_policy='propagate'])
#又称"标准分数"(Standard Score)
求"相对Z分数"(Relative Z-Scores):[<zscore>=]scipy.stats.zmap(<scores>,<compare>[,axis=0,ddof=0])