假设我有一个频率为4个可能的事件的样本:
Event1 - 5
E2 - 1
E3 - 0
E4 - 12
并且我具有发生事件的预期概率:
p1 - 0.2
p2 - 0.1
p3 - 0.1
p4 - 0.6
利用我四个事件的观测频率之和(18),我可以计算事件的预期频率,对吗?
expectedE1 - 18 * 0.2 = 3.6
expectedE2 - 18 * 0.1 = 1.8
expectedE1 - 18 * 0.1 = 1.8
expectedE1 - 18 * 0.6 = 10.8
如何比较观察值与期望值?测试我计算的概率是否是好的预测因子?
我想到了卡方检验,但是结果随样本大小变化(n = 18),我的意思是,如果将观察值乘以1342,并使用相同的方法,结果将有所不同。也许wilcox配对测试有效,但是您有何建议?
如果可以在R中建议,那会更好。
r
statistical-significance
chi-squared
multivariate-analysis
exponential
joint-distribution
statistical-significance
self-study
standard-deviation
probability
normal-distribution
spss
interpretation
assumptions
cox-model
reporting
cox-model
statistical-significance
reliability
method-comparison
classification
boosting
ensemble
adaboost
confidence-interval
cross-validation
prediction
prediction-interval
regression
machine-learning
svm
regularization
regression
sampling
survey
probit
matlab
feature-selection
information-theory
mutual-information
time-series
forecasting
simulation
classification
boosting
ensemble
adaboost
normal-distribution
multivariate-analysis
covariance
gini
clustering
text-mining
distance-functions
information-retrieval
similarities
regression
logistic
stata
group-differences
r
anova
confidence-interval
repeated-measures
r
logistic
lme4-nlme
inference
fiducial
kalman-filter
classification
discriminant-analysis
linear-algebra
computing
statistical-significance
time-series
panel-data
missing-data
uncertainty
probability
multivariate-analysis
r
classification
spss
k-means
discriminant-analysis
poisson-distribution
average
r
random-forest
importance
probability
conditional-probability
distributions
standard-deviation
time-series
machine-learning
online
forecasting
r
pca
dataset
data-visualization
bayes
distributions
mathematical-statistics
degrees-of-freedom