Perguntas com a marcação «wilcoxon-signed-rank»

O teste de classificação sinalizada de Wilcoxon é um teste de classificação não paramétrico para comparar duas amostras pareadas, se os valores em uma são maiores do que na outra. Também pode ser usado para comparar uma amostra a um valor fixo. [O teste NÃO deve ser confundido com o `teste de sinal`].

29
Como lidar com dados hierárquicos / aninhados no aprendizado de máquina

Vou explicar meu problema com um exemplo. Suponha que você queira prever a renda de um indivíduo, com alguns atributos: {Idade, Sexo, País, Região, Cidade}. Você tem um conjunto de dados de treinamento como esse train <- data.frame(CountryID=c(1,1,1,1, 2,2,2,2, 3,3,3,3), RegionID=c(1,1,1,2,...

regression machine-learning multilevel-analysis correlation dataset spatial paired-comparisons cross-correlation clustering aic bic dependent-variable k-means mean standard-error measurement-error errors-in-variables regression multiple-regression pca linear-model dimensionality-reduction machine-learning neural-networks deep-learning conv-neural-network computer-vision clustering spss r weighted-data wilcoxon-signed-rank bayesian hierarchical-bayesian bugs stan distributions categorical-data variance ecology r survival regression r-squared descriptive-statistics cross-section maximum-likelihood factor-analysis likert r multiple-imputation propensity-scores distributions t-test logit probit z-test confidence-interval poisson-distribution deep-learning conv-neural-network residual-networks r survey wilcoxon-mann-whitney ranking kruskal-wallis bias loss-functions frequentist decision-theory risk machine-learning distributions normal-distribution multivariate-analysis inference dataset factor-analysis survey multilevel-analysis clinical-trials
15
Qual é a intuição por trás de amostras intercambiáveis ​​sob a hipótese nula?

Os testes de permutação (também chamados de teste de randomização, teste de re-randomização ou teste exato) são muito úteis e úteis quando a suposição de distribuição normal exigida por, por exemplo, t-testnão é atendida e quando a transformação dos valores pela classificação do teste...

hypothesis-testing permutation-test exchangeability r statistical-significance loess data-visualization normal-distribution pdf ggplot2 kernel-smoothing probability self-study expected-value normal-distribution prior correlation time-series regression heteroscedasticity estimation estimators fisher-information data-visualization repeated-measures binary-data panel-data mathematical-statistics coefficient-of-variation normal-distribution order-statistics regression machine-learning one-class probability estimators forecasting prediction validation finance measurement-error variance mean spatial monte-carlo data-visualization boxplot sampling uniform chi-squared goodness-of-fit probability mixture theory gaussian-mixture regression statistical-significance p-value bootstrap regression multicollinearity correlation r poisson-distribution survival regression categorical-data ordinal-data ordered-logit regression interaction time-series machine-learning forecasting cross-validation binomial multiple-comparisons simulation false-discovery-rate r clustering frequency wilcoxon-mann-whitney wilcoxon-signed-rank r svm t-test missing-data excel r numerical-integration r random-variable lme4-nlme mixed-model weighted-regression power-law errors-in-variables machine-learning classification entropy information-theory mutual-information
14
Caret glmnet vs cv.glmnet

Parece haver muita confusão na comparação entre usar glmnetdentro caretpara procurar uma lambda ideal e usar cv.glmnetpara fazer a mesma tarefa. Muitas perguntas foram feitas, por exemplo: Modelo de classificação train.glmnet vs. cv.glmnet? Qual é a maneira correta de usar glmnet com...

r caret glmnet machine-learning neural-networks maximum softmax probability distributions mathematical-statistics random-variable cdf statistical-significance variance expected-value ratio sample-size reliability tolerance-interval wilcoxon-signed-rank self-study variance sampling mean machine-learning svm libsvm self-study sampling ranks data-visualization histogram machine-learning classification normal-distribution mathematical-statistics maximum-likelihood mixture predictive-models prediction seasonality
10
Modelo de Histórico de Eventos em Tempo Discreto (Sobrevivência) em R

Estou tentando ajustar um modelo de tempo discreto no R, mas não sei como fazê-lo. Eu li que você pode organizar a variável dependente em linhas diferentes, uma para cada observação no tempo, e usar a glmfunção com um link logit ou cloglog. Neste sentido, tem três colunas: ID, Event(1 ou 0, em...

r survival pca sas matlab neural-networks r logistic spatial spatial-interaction-model r time-series econometrics var statistical-significance t-test cross-validation sample-size r regression optimization least-squares constrained-regression nonparametric ordinal-data wilcoxon-signed-rank references neural-networks jags bugs hierarchical-bayesian gaussian-mixture r regression svm predictive-models libsvm scikit-learn probability self-study stata sample-size spss wilcoxon-mann-whitney survey ordinal-data likert group-differences r regression anova mathematical-statistics normal-distribution random-generation truncation repeated-measures variance variability distributions random-generation uniform regression r generalized-linear-model goodness-of-fit data-visualization r time-series arima autoregressive confidence-interval r time-series arima autocorrelation seasonality hypothesis-testing bayesian frequentist uninformative-prior correlation matlab cross-correlation
10
Por que Anova () e drop1 () forneceram respostas diferentes para os GLMMs?

Eu tenho um GLMM do formulário: lmer(present? ~ factor1 + factor2 + continuous + factor1*continuous + (1 | factor3), family=binomial) Quando uso drop1(model, test="Chi"), obtenho resultados diferentes dos que utilizo Anova(model, type="III")na embalagem do carro ou summary(model). Estes dois...

r anova glmm r mixed-model bootstrap sample-size cross-validation roc auc sampling stratification random-allocation logistic stata interpretation proportion r regression multiple-regression linear-model lm r cross-validation cart rpart logistic generalized-linear-model econometrics experiment-design causality instrumental-variables random-allocation predictive-models data-mining estimation contingency-tables epidemiology standard-deviation mean ancova psychology statistical-significance cross-validation synthetic-data poisson-distribution negative-binomial bioinformatics sequence-analysis distributions binomial classification k-means distance unsupervised-learning euclidean correlation chi-squared spearman-rho forecasting excel exponential-smoothing binomial sample-size r change-point wilcoxon-signed-rank ranks clustering matlab covariance covariance-matrix normal-distribution simulation random-generation bivariate standardization confounding z-statistic forecasting arima minitab poisson-distribution negative-binomial poisson-regression overdispersion probability self-study markov-process estimation maximum-likelihood classification pca group-differences chi-squared survival missing-data contingency-tables anova proportion
8
Teste post hoc em uma ANOVA de design misto 2x3 usando SPSS?

Eu tenho dois grupos de 10 participantes que foram avaliados três vezes durante um experimento. Para testar as diferenças entre os grupos e nas três avaliações, executei um ANOVA de desenho misto 2x3 com group(controle, experimental), time(primeiro, segundo, três) e group x time. Ambos timee...

anova mixed-model spss post-hoc bonferroni time-series unevenly-spaced-time-series classification normal-distribution discriminant-analysis probability normal-distribution estimation sampling classification svm terminology pivot-table random-generation self-study estimation sampling estimation categorical-data maximum-likelihood excel least-squares instrumental-variables 2sls total-least-squares correlation self-study variance unbiased-estimator bayesian mixed-model ancova statistical-significance references p-value fishers-exact probability monte-carlo particle-filter logistic predictive-models modeling interaction survey hypothesis-testing multiple-regression regression variance data-transformation residuals minitab r time-series forecasting arima garch correlation estimation least-squares bias pca predictive-models genetics sem partial-least-squares nonparametric ordinal-data wilcoxon-mann-whitney bonferroni wilcoxon-signed-rank traminer regression econometrics standard-error robust misspecification r probability logistic generalized-linear-model r-squared effect-size gee ordered-logit bayesian classification svm kernel-trick nonlinear bayesian pca dimensionality-reduction eigenvalues probability distributions mathematical-statistics estimation nonparametric kernel-smoothing expected-value filter mse time-series correlation data-visualization clustering estimation predictive-models recommender-system sparse hypothesis-testing data-transformation parametric probability summations correlation pearson-r spearman-rho bayesian replicability dimensionality-reduction discriminant-analysis outliers weka
8
Encontre distribuição e transforme em distribuição normal

Eu tenho dados que descrevem com que frequência um evento ocorre durante uma hora ("número por hora", nph) e quanto tempo os eventos duram ("duração em segundos por hora", dph). Estes são os dados originais: nph <- c(2.50000000003638, 3.78947368414551, 1.51456310682008, 5.84686774940732,...

normal-distribution data-transformation logistic generalized-linear-model ridge-regression t-test wilcoxon-signed-rank paired-data naive-bayes distributions logistic goodness-of-fit time-series eviews ecm panel-data reliability psychometrics validity cronbachs-alpha self-study random-variable expected-value median regression self-study multiple-regression linear-model forecasting prediction-interval normal-distribution excel bayesian multivariate-analysis modeling predictive-models canonical-correlation rbm time-series machine-learning neural-networks fishers-exact factorisation-theorem svm prediction linear reinforcement-learning cdf probability-inequalities ecdf time-series kalman-filter state-space-models dynamic-regression index-decomposition sampling stratification cluster-sample survey-sampling distributions maximum-likelihood gamma-distribution