Perguntas com a marcação «data-mining»

11
R / mgcv: Por que os produtos tensores te () e ti () produzem superfícies diferentes?

O mgcvpacote para Rpossui duas funções para ajustar as interações do produto tensorial: te()e ti(). Entendo a divisão básica do trabalho entre os dois (ajustando uma interação não linear versus decompondo essa interação em efeitos principais e uma interação). O que não entendo é o porquê te(x1,...

r gam mgcv conditional-probability mixed-model references bayesian estimation conditional-probability machine-learning optimization gradient-descent r hypothesis-testing wilcoxon-mann-whitney time-series bayesian inference change-point time-series anova repeated-measures statistical-significance bayesian contingency-tables regression prediction quantiles classification auc k-means scikit-learn regression spatial circular-statistics t-test effect-size cohens-d r cross-validation feature-selection caret machine-learning modeling python optimization frequentist correlation sample-size normalization group-differences heteroscedasticity independence generalized-least-squares lme4-nlme references mcmc metropolis-hastings optimization r logistic feature-selection separation clustering k-means normal-distribution gaussian-mixture kullback-leibler java spark-mllib data-visualization categorical-data barplot hypothesis-testing statistical-significance chi-squared type-i-and-ii-errors pca scikit-learn conditional-expectation statistical-significance meta-analysis intuition r time-series multivariate-analysis garch machine-learning classification data-mining missing-data cart regression cross-validation matrix-decomposition categorical-data repeated-measures chi-squared assumptions contingency-tables prediction binary-data trend test-for-trend matrix-inverse anova categorical-data regression-coefficients standard-error r distributions exponential interarrival-time copula log-likelihood time-series forecasting prediction-interval mean standard-error meta-analysis meta-regression network-meta-analysis systematic-review normal-distribution multiple-regression generalized-linear-model poisson-distribution poisson-regression r sas cohens-kappa
10
Como posso agrupar strings por temas comuns?

Estou tentando agrupar, por exemplo, strings sobre programação com outras strings sobre programação, strings sobre física com outras strings sobre física, etc., para uma ampla gama de tópicos. Apesar do aspecto linguístico teórico gritante do problema, estou procurando fazer isso usando programação...

natural-language data-mining
10
Variável categórica de regressão linear R valor "oculto"

Este é apenas um exemplo que encontrei várias vezes, portanto não tenho dados de amostra. Executando um modelo de regressão linear em R: a.lm = lm(Y ~ x1 + x2) x1é uma variável contínua. x2é categórico e possui três valores, por exemplo, "Baixo", "Médio" e "Alto". No entanto, a saída fornecida...

r regression categorical-data regression-coefficients categorical-encoding machine-learning random-forest anova spss r self-study bootstrap monte-carlo r multiple-regression partitioning neural-networks normalization machine-learning svm kernel-trick self-study survival cox-model repeated-measures survey likert correlation variance sampling meta-analysis anova independence sample assumptions bayesian covariance r regression time-series mathematical-statistics graphical-model machine-learning linear-model kernel-trick linear-algebra self-study moments function correlation spss probability confidence-interval sampling mean population r generalized-linear-model prediction offset data-visualization clustering sas cart binning sas logistic causality regression self-study standard-error r distributions r regression time-series multiple-regression python chi-squared independence sample clustering data-mining rapidminer probability stochastic-processes clustering binary-data dimensionality-reduction svd correspondence-analysis data-visualization excel c# hypothesis-testing econometrics survey rating composite regression least-squares mcmc markov-process kullback-leibler convergence predictive-models r regression anova confidence-interval survival cox-model hazard normal-distribution autoregressive mixed-model r mixed-model sas hypothesis-testing mediation interaction
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