bruceR 简单应用

最新推荐文章于 2024-08-29 20:01:42 发布
最新推荐文章于 2024-08-29 20:01:42 发布
阅读量1.5k 收藏 16
点赞数 4
分类专栏： R
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/nikang3148/article/details/116903018
版权

bruceR

bruceR包中函数相对基本函数提供了更多的模型输出参数，如效应值等。

加载bruceR

library(bruceR)
## Warning: package 'bruceR' was built under R version 4.0.5
## 
## <U+0001F381> bruceR: BRoadly Useful Convenient and Efficient R functions
## 
## Loaded R packages:
## [Data]: rio / dplyr / tidyr / stringr / forcats / data.table
## [Stat]: psych / emmeans / effectsize / performance
## [Plot]: ggplot2 / ggtext / cowplot / see
## 
## Frequently used functions in `bruceR`:
## set.wd() / Describe() / Freq() / Corr() / Alpha() / MEAN()
## MANOVA() / EMMEANS() / model_summary() / theme_bruce()

描述统计

Describe(iris,all.as.numeric = F,plot=T,upper.triangle = T,upper.smooth = "lm")
## Descriptive Statistics:
## ────────────────────────────────────────────────────────────────
##                 N Mean   SD | Median  Min  Max Skewness Kurtosis
## ────────────────────────────────────────────────────────────────
## Sepal.Length  150 5.84 0.83 |   5.80 4.30 7.90     0.31    -0.61
## Sepal.Width   150 3.06 0.44 |   3.00 2.00 4.40     0.31     0.14
## Petal.Length  150 3.76 1.77 |   4.35 1.00 6.90    -0.27    -1.42
## Petal.Width   150 1.20 0.76 |   1.30 0.10 2.50    -0.10    -1.36
## Species*      150 2.00 0.82 |   2.00 1.00 3.00     0.00    -1.52
## ────────────────────────────────────────────────────────────────
## Registered S3 method overwritten by 'GGally':
##   method from   
##   +.gg   ggplot2
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

相关分析

iris_cor <- iris %>% select(-Species) %>%
  Corr(method="pearson",p.adjust="fdr",plot.palette = c("red", "white", "blue"))
## Correlation matrix is displayed in plot.
## p values ABOVE the diagonal are adjusted using the "fdr" method.

## Pearson's r and 95% confidence intervals:
## p values and 95% CIs are adjusted using the "fdr" method.
## ─────────────────────────────────────────────────────────────
##                                r       [95% CI]     p       N
## ─────────────────────────────────────────────────────────────
## Sepal.Length-Sepal.Width   -0.12 [-0.32,  0.10]  .152     150
## Sepal.Length-Petal.Length   0.87 [ 0.81,  0.92] <.001 *** 150
## Sepal.Length-Petal.Width    0.82 [ 0.73,  0.88] <.001 *** 150
## Sepal.Width-Petal.Length   -0.43 [-0.59, -0.24] <.001 *** 150
## Sepal.Width-Petal.Width    -0.37 [-0.54, -0.16] <.001 *** 150
## Petal.Length-Petal.Width    0.96 [ 0.94,  0.98] <.001 *** 150
## ─────────────────────────────────────────────────────────────

相关系数差异比较

###独立检验 Sepal.Length~Petal.Length vs Petal.Width~Sepal.Width
cor_diff(r1=0.87, n1=150, r2=-0.37, n2=150)
## r1 = 0.870 (N = 150)
## r2 = -0.370 (N = 150)
## Difference of correlation: z = 14.76, p = 3e-49 ***
### 非独立检验Sepal.Length~Petal.Length vs Petal.Width~Petal.Length
cor_diff(r1=0.87, r2=0.96, n=150,rcov=0.82)
## r1 = 0.870
## r2 = 0.960
## (N = 150, r_cov = 0.820)
## Difference of correlation: t(147) = -7.14, p = 4e-11 ***

交叉相关分析

# 用于分析时间序列数据，以发现是否存在时间延迟
ccf_plot(Petal.Length~Sepal.Width,data=iris)

频率分析

Freq(iris$Species)
## Frequency Statistics:
## ───────────────────
##              N    %
## ───────────────────
## setosa      50 33.3
## versicolor  50 33.3
## virginica   50 33.3
## ───────────────────
## Total N = 150
Freq(iris$Sepal.Length,sort="+")
## Frequency Statistics:
## ───────────
##       N   %
## ───────────
## 4.3   1 0.7
## 4.5   1 0.7
## 5.3   1 0.7
## 7     1 0.7
## 7.1   1 0.7
## 7.3   1 0.7
## 7.4   1 0.7
## 7.6   1 0.7
## 7.9   1 0.7
## 4.7   2 1.3
## 6.6   2 1.3
## 4.4   3 2.0
## 5.9   3 2.0
## 6.8   3 2.0
## 7.2   3 2.0
## 4.6   4 2.7
## 5.2   4 2.7
## 6.2   4 2.7
## 6.9   4 2.7
## 7.7   4 2.7
## 4.8   5 3.3
## 6.5   5 3.3
## 4.9   6 4.0
## 5.4   6 4.0
## 5.6   6 4.0
## 6     6 4.0
## 6.1   6 4.0
## 5.5   7 4.7
## 5.8   7 4.7
## 6.4   7 4.7
## 5.7   8 5.3
## 6.7   8 5.3
## 5.1   9 6.0
## 6.3   9 6.0
## 5    10 6.7
## ───────────
## Total N = 150

回归分析

mtcars<-as_tibble(mtcars)
mtcars
## # A tibble: 32 x 11
##      mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
##    <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
##  1  21       6  160    110  3.9   2.62  16.5     0     1     4     4
##  2  21       6  160    110  3.9   2.88  17.0     0     1     4     4
##  3  22.8     4  108     93  3.85  2.32  18.6     1     1     4     1
##  4  21.4     6  258    110  3.08  3.22  19.4     1     0     3     1
##  5  18.7     8  360    175  3.15  3.44  17.0     0     0     3     2
##  6  18.1     6  225    105  2.76  3.46  20.2     1     0     3     1
##  7  14.3     8  360    245  3.21  3.57  15.8     0     0     3     4
##  8  24.4     4  147.    62  3.69  3.19  20       1     0     4     2
##  9  22.8     4  141.    95  3.92  3.15  22.9     1     0     4     2
## 10  19.2     6  168.   123  3.92  3.44  18.3     1     0     4     4
## # ... with 22 more rows

简单线性回归

lm(mpg~wt*disp,data=mtcars)%>%
  GLM_summary()
## MODEL INFO:
## Model type: General Linear Model (GLM) (OLS Regression)
## Observations: N = 32
## 
## MODEL FIT:
## F(3, 28) = 52.95, p = 1e-11 ***
## R2 = 0.85014 (Adjusted R2 = 0.83408)
## 
## FIXED EFFECTS:
## Outcome variable: mpg (N = 32)
## ───────────────────────────────────────────────────────────────────
##                   b    S.E.     t     p        [95% CI of b]    VIF
## ───────────────────────────────────────────────────────────────────
## (Intercept)  44.082 (3.123) 14.11 <.001 *** [37.685, 50.479]       
## wt           -6.496 (1.313) -4.95 <.001 *** [-9.186, -3.805]  8.495
## disp         -0.056 (0.013) -4.26 <.001 *** [-0.083, -0.029] 13.848
## wt:disp       0.012 (0.003)  3.60  .001 **  [ 0.005,  0.018] 24.294
## ───────────────────────────────────────────────────────────────────
## 
## Standardized coefficients: mpg (N = 32)
## ─────────────────────────────────────────────────────────────
##           Beta*   S.E.*     [95% CI of Beta] r.partial r.part
## ─────────────────────────────────────────────────────────────
## wt       -1.055 (0.213) *** [-1.491, -0.618]    -0.683 -0.362
## disp     -1.159 (0.272) *** [-1.717, -0.601]    -0.627 -0.311
## wt:disp   1.297 (0.361) **  [ 0.558,  2.035]     0.562  0.263
## ─────────────────────────────────────────────────────────────
lm(mpg~wt*disp,data=mtcars)%>%
  model_summary(std_coef=F)
## 
## =======================
##              (1) mpg   
## -----------------------
## (Intercept)  44.082 ***
##              (3.123)   
## wt           -6.496 ***
##              (1.313)   
## disp         -0.056 ***
##              (0.013)   
## wt:disp       0.012 ** 
##              (0.003)   
## -----------------------
## R^2           0.850    
## Adj. R^2      0.834    
## Num. obs.    32        
## =======================
## Note. * p < .05, ** p < .01, *** p < .001.
## 
## # Check for Multicollinearity
## 
## Moderate Correlation
## 
##  Term  VIF Increased SE Tolerance
##    wt 8.49         2.91      0.12
## 
## High Correlation
## 
##     Term   VIF Increased SE Tolerance
##     disp 13.85         3.72      0.07
##  wt:disp 24.29         4.93      0.04
# one step
regress(mpg ~ wt*disp, data=mtcars, robust=F)
## MODEL INFO:
## Model type: General Linear Model (GLM) (OLS Regression)
## Observations: N = 32
## 
## MODEL FIT:
## F(3, 28) = 52.95, p = 1e-11 ***
## R2 = 0.85014 (Adjusted R2 = 0.83408)
## 
## FIXED EFFECTS:
## Outcome variable: mpg (N = 32)
## ───────────────────────────────────────────────────────────────────
##                   b    S.E.     t     p        [95% CI of b]    VIF
## ───────────────────────────────────────────────────────────────────
## (Intercept)  44.082 (3.123) 14.11 <.001 *** [37.685, 50.479]       
## wt           -6.496 (1.313) -4.95 <.001 *** [-9.186, -3.805]  8.495
## disp         -0.056 (0.013) -4.26 <.001 *** [-0.083, -0.029] 13.848
## wt:disp       0.012 (0.003)  3.60  .001 **  [ 0.005,  0.018] 24.294
## ───────────────────────────────────────────────────────────────────
## 
## Standardized coefficients: mpg (N = 32)
## ─────────────────────────────────────────────────────────────
##           Beta*   S.E.*     [95% CI of Beta] r.partial r.part
## ─────────────────────────────────────────────────────────────
## wt       -1.055 (0.213) *** [-1.491, -0.618]    -0.683 -0.362
## disp     -1.159 (0.272) *** [-1.717, -0.601]    -0.627 -0.311
## wt:disp   1.297 (0.361) **  [ 0.558,  2.035]     0.562  0.263
## ─────────────────────────────────────────────────────────────
# tidy style
lm(mpg~wt*disp,data=mtcars)%>%broom::tidy()
## Registered S3 methods overwritten by 'broom':
##   method            from  
##   tidy.glht         jtools
##   tidy.summary.glht jtools
## # A tibble: 4 x 5
##   term        estimate std.error statistic  p.value
##   <chr>          <dbl>     <dbl>     <dbl>    <dbl>
## 1 (Intercept)  44.1      3.12        14.1  2.96e-14
## 2 wt           -6.50     1.31        -4.95 3.22e- 5
## 3 disp         -0.0564   0.0132      -4.26 2.10e- 4
## 4 wt:disp       0.0117   0.00326      3.60 1.23e- 3
lm(mpg~wt*disp,data=mtcars)%>%broom::glance()
## # A tibble: 1 x 12
##   r.squared adj.r.squared sigma statistic  p.value    df logLik   AIC   BIC
##       <dbl>         <dbl> <dbl>     <dbl>    <dbl> <dbl>  <dbl> <dbl> <dbl>
## 1     0.850         0.834  2.45      52.9 1.16e-11     3  -72.0  154.  161.
## # ... with 3 more variables: deviance <dbl>, df.residual <int>, nobs <int>

混合模型

regress(mpg ~ wt*disp+(1|cyl), data=mtcars, robust=F)
## boundary (singular) fit: see ?isSingular
## MODEL INFO:
## Model type: Linear Mixed Model (LMM)
## = Hierarchical Linear Model (HLM)
## = Multilevel Linear Model (MLM)
## 
## Formula: mpg ~ wt * disp + (1 | cyl)
## Level-2 predictors: ''
## 
##             Variable Type
## (Intercept)     Intercept
## wt                L1fixed
## disp              L1fixed
## wt:disp           L1fixed
## 
## Level-1 Observations: N = 32
## Level-2 Groups/Clusters: cyl, 3
## 
## MODEL FIT:
## AIC = 173.382
## BIC = 182.176
## R_(m)2 = 0.83671  (Marginal R2: fixed effects)
## R_(c)2 = 0.83671  (Conditional R2: fixed + random effects)
## Omega2 = 0.85014  (= 1 - proportion of unexplained variance)
## 
## ANOVA table:
## ───────────────────────────────────────────────────
##          Sum Sq Mean Sq NumDF DenDF     F     p    
## ───────────────────────────────────────────────────
## wt       147.42  147.42     1 28.00 24.46 <.001 ***
## disp     109.22  109.22     1 28.00 18.12 <.001 ***
## wt:disp   77.93   77.93     1 28.00 12.93  .001 ** 
## ───────────────────────────────────────────────────
## 
## FIXED EFFECTS:
## Outcome variable: mpg
## ─────────────────────────────────────────────────────────────────
##                   b    S.E.     t   df     p        [95% CI of b]
## ─────────────────────────────────────────────────────────────────
## (Intercept)  44.082 (3.123) 14.11 28.0 <.001 *** [37.685, 50.479]
## wt           -6.496 (1.313) -4.95 28.0 <.001 *** [-9.186, -3.805]
## disp         -0.056 (0.013) -4.26 28.0 <.001 *** [-0.083, -0.029]
## wt:disp       0.012 (0.003)  3.60 28.0  .001 **  [ 0.005,  0.018]
## ─────────────────────────────────────────────────────────────────
## 'df' is estimated by Satterthwaite approximation.
## 
## Standardized coefficients: mpg
## ────────────────────────────────────────────────────────────
##           Beta*   S.E.*    t* df*    p*     [95% CI of Beta]
## ────────────────────────────────────────────────────────────
## wt       -1.055 (0.213) -4.95  28 <.001 *** [-1.491, -0.618]
## disp     -1.159 (0.272) -4.26  28 <.001 *** [-1.717, -0.601]
## wt:disp   1.297 (0.361)  3.60  28  .001 **  [ 0.558,  2.035]
## ────────────────────────────────────────────────────────────
## 'df*' is calculated based on variable types.
## 
## RANDOM EFFECTS:
## ──────────────────────────────────────────
##  Cluster  K   Parameter   Variance     ICC
## ──────────────────────────────────────────
##  cyl        3 (Intercept)  0.00000 0.00000
##  Residual                  6.02673        
## ──────────────────────────────────────────

方差分析

单因素组间方差分析 one-way ANOVA

iris %>% MANOVA(dv="Sepal.Width", between=c("Species")) %>%
  EMMEANS("Species")
## ====== MANOVA Output (Between-Subjects Design) ======
## 
## Descriptive Statistics:
## ────────────────────────
##     Species    M   SD  n
## ────────────────────────
##  setosa     3.43 0.38 50
##  versicolor 2.77 0.31 50
##  virginica  2.97 0.32 50
## ────────────────────────
## Total sample size: N = 150
## 
## ANOVA Table:
## Dependent variable(s):      Sepal.Width
## Between-subjects factor(s): Species
## Within-subjects factor(s):  –
## Covariate(s):               –
## ────────────────────────────────────────────────────────────
##            MS  MSE df1 df2     F     p        η2p [90% CI]
## ────────────────────────────────────────────────────────────
## Species  5.67 0.12   2 147 49.16 <.001 *** .401 [.300, .485]
## ────────────────────────────────────────────────────────────
## MSE = Mean Square Error (an estimate of population variance σ2).
## 
## ω2 = omega-squared = (SS - df1 * MSE) / (SST + MSE)
## η2 = eta-squared = SS / SST
## η2G = generalized eta-squared (see Olejnik & Algina, 2003)
## η2p = partial eta-squared = SS / (SS + SSE) = F * df1 / (F * df1 + df2)
## Cohen’s f = sqrt( η2p / (1 - η2p) )
## 
## Levene’s Test for Homogeneity of Variance:
## DV = Sepal.Width:
##                 Levene's F df1 df2     p sig
## Based on Mean         0.60   2 147  .550    
## Based on Median       0.59   2 147  .556    
## 
## ------ EMMEANS Output (effect = "Species") ------
## 
## Omnibus Test of "Species":
## ─────────────────────────────────────────────────────
##   Effect df1 df2     F     p           η2p [90% CI]
## ─────────────────────────────────────────────────────
##  Species   2 147 49.16 <.001 *** 0.401 [0.300, 0.485]
## ─────────────────────────────────────────────────────
## 
## Estimated Marginal Means of "Species":
##  Species    Mean S.E.   [95% CI of Mean]
##  setosa     3.43 (0.05) [3.33, 3.52]    
##  versicolor 2.77 (0.05) [2.68, 2.86]    
##  virginica  2.97 (0.05) [2.88, 3.07]    
## 
## Estimated means use an equally weighted average. 
## 
## Pairwise Comparisons of "Species":
##  Contrast               b     S.E.    df t     p*        Cohen's d [95% CI]  
##  versicolor - setosa    -0.66 (0.07) 147 -9.69 <.001 *** -1.94 [-2.42, -1.45]
##  virginica - setosa     -0.45 (0.07) 147 -6.68 <.001 *** -1.34 [-1.82, -0.85]
##  virginica - versicolor  0.20 (0.07) 147  3.00  .009 **   0.60 [ 0.12,  1.08]
## 
## SD_pooled for computing Cohen’s d: 0.34 
## P-value adjustment: Bonferroni method for 3 tests. 
## Disclaimer (about Cohen’s d):
## There is considerable disagreement on how to compute Cohen’s d.
## You should not take the above output as the only right results.
## You are completely responsible for setting the "sd.pooled".

二因素方差分析

# 混合分析
MANOVA(data=CO2,
       dv="uptake",
       between=c("Type","Treatment")) %>%
  EMMEANS(c("Type","Treatment"))
## ====== MANOVA Output (Between-Subjects Design) ======
## 
## Descriptive Statistics:
## ─────────────────────────────────────
##         Type  Treatment     M   SD  n
## ─────────────────────────────────────
##  Quebec      nonchilled 35.33 9.60 21
##  Quebec      chilled    31.75 9.64 21
##  Mississippi nonchilled 25.95 7.40 21
##  Mississippi chilled    15.81 4.06 21
## ─────────────────────────────────────
## Total sample size: N = 84
## 
## ANOVA Table:
## Dependent variable(s):      uptake
## Between-subjects factor(s): Type, Treatment
## Within-subjects factor(s):  –
## Covariate(s):               –
## ───────────────────────────────────────────────────────────────────────
##                      MS   MSE df1 df2     F     p        η2p [90% CI]
## ───────────────────────────────────────────────────────────────────────
## Type            3365.53 64.09   1  80 52.51 <.001 *** .396 [.262, .510]
## Treatment        988.11 64.09   1  80 15.42 <.001 *** .162 [.057, .284]
## Type:Treatment   225.73 64.09   1  80  3.52  .064 .   .042 [.000, .135]
## ───────────────────────────────────────────────────────────────────────
## MSE = Mean Square Error (an estimate of population variance σ2).
## 
## ω2 = omega-squared = (SS - df1 * MSE) / (SST + MSE)
## η2 = eta-squared = SS / SST
## η2G = generalized eta-squared (see Olejnik & Algina, 2003)
## η2p = partial eta-squared = SS / (SS + SSE) = F * df1 / (F * df1 + df2)
## Cohen’s f = sqrt( η2p / (1 - η2p) )
## 
## Levene’s Test for Homogeneity of Variance:
## DV = uptake:
##                 Levene's F df1 df2     p sig
## Based on Mean         3.36   3  80  .023 *  
## Based on Median       1.50   3  80  .221    
## 
## ------ EMMEANS Output (effect = "Type" & "Treatment") ------
## 
## Omnibus Test of "Type" & "Treatment":
## ────────────────────────────────────────────────────────────
##          Effect df1 df2     F     p           η2p [90% CI]
## ────────────────────────────────────────────────────────────
##  Type             1  80 52.51 <.001 *** 0.396 [0.262, 0.510]
##  Treatment        1  80 15.42 <.001 *** 0.162 [0.057, 0.284]
##  Type:Treatment   1  80  3.52  .064 .   0.042 [0.000, 0.135]
## ────────────────────────────────────────────────────────────
## 
## Estimated Marginal Means of "Type" & "Treatment":
##  Type        Treatment  Mean  S.E.   [95% CI of Mean]
##  Quebec      nonchilled 35.33 (1.75) [31.86, 38.81]  
##  Mississippi nonchilled 25.95 (1.75) [22.48, 29.43]  
##  Quebec      chilled    31.75 (1.75) [28.28, 35.23]  
##  Mississippi chilled    15.81 (1.75) [12.34, 19.29]  
## 
## Estimated means use an equally weighted average. 
## 
## Pairwise Comparisons of "Type" & "Treatment":
##  Contrast                                     b      S.E.   df t     p*       
##  Mississippi nonchilled - Quebec nonchilled    -9.38 (2.47) 80 -3.80  .002 ** 
##  Quebec chilled - Quebec nonchilled            -3.58 (2.47) 80 -1.45  .907    
##  Quebec chilled - Mississippi nonchilled        5.80 (2.47) 80  2.35  .128    
##  Mississippi chilled - Quebec nonchilled      -19.52 (2.47) 80 -7.90 <.001 ***
##  Mississippi chilled - Mississippi nonchilled -10.14 (2.47) 80 -4.10 <.001 ***
##  Mississippi chilled - Quebec chilled         -15.94 (2.47) 80 -6.45 <.001 ***
##  Cohen's d [95% CI]  
##  -1.17 [-2.01, -0.34]
##  -0.45 [-1.28,  0.39]
##   0.72 [-0.11,  1.56]
##  -2.44 [-3.27, -1.60]
##  -1.27 [-2.10, -0.43]
##  -1.99 [-2.83, -1.16]
## 
## SD_pooled for computing Cohen’s d: 8.01 
## P-value adjustment: Bonferroni method for 6 tests. 
## Disclaimer (about Cohen’s d):
## There is considerable disagreement on how to compute Cohen’s d.
## You should not take the above output as the only right results.
## You are completely responsible for setting the "sd.pooled".
# 分组简单效应分析，by
MANOVA(data=CO2,
       dv="uptake",
       between=c("Type","Treatment"))%>%
  EMMEANS("Type",by="Treatment")
## ====== MANOVA Output (Between-Subjects Design) ======
## 
## Descriptive Statistics:
## ─────────────────────────────────────
##         Type  Treatment     M   SD  n
## ─────────────────────────────────────
##  Quebec      nonchilled 35.33 9.60 21
##  Quebec      chilled    31.75 9.64 21
##  Mississippi nonchilled 25.95 7.40 21
##  Mississippi chilled    15.81 4.06 21
## ─────────────────────────────────────
## Total sample size: N = 84
## 
## ANOVA Table:
## Dependent variable(s):      uptake
## Between-subjects factor(s): Type, Treatment
## Within-subjects factor(s):  –
## Covariate(s):               –
## ───────────────────────────────────────────────────────────────────────
##                      MS   MSE df1 df2     F     p        η2p [90% CI]
## ───────────────────────────────────────────────────────────────────────
## Type            3365.53 64.09   1  80 52.51 <.001 *** .396 [.262, .510]
## Treatment        988.11 64.09   1  80 15.42 <.001 *** .162 [.057, .284]
## Type:Treatment   225.73 64.09   1  80  3.52  .064 .   .042 [.000, .135]
## ───────────────────────────────────────────────────────────────────────
## MSE = Mean Square Error (an estimate of population variance σ2).
## 
## ω2 = omega-squared = (SS - df1 * MSE) / (SST + MSE)
## η2 = eta-squared = SS / SST
## η2G = generalized eta-squared (see Olejnik & Algina, 2003)
## η2p = partial eta-squared = SS / (SS + SSE) = F * df1 / (F * df1 + df2)
## Cohen’s f = sqrt( η2p / (1 - η2p) )
## 
## Levene’s Test for Homogeneity of Variance:
## DV = uptake:
##                 Levene's F df1 df2     p sig
## Based on Mean         3.36   3  80  .023 *  
## Based on Median       1.50   3  80  .221    
## 
## ------ EMMEANS Output (effect = "Type") ------
## 
## Simple Effects of "Type":
## ─────────────────────────────────────────────────────────────────────────────
##  Effect          (By: Treatment) df1 df2     F     p           η2p [90% CI]
## ─────────────────────────────────────────────────────────────────────────────
##    Type (Treatment = nonchilled)   1  80 14.42 <.001 *** 0.153 [0.051, 0.275]
##    Type (Treatment = chilled)      1  80 41.61 <.001 *** 0.342 [0.208, 0.461]
## ─────────────────────────────────────────────────────────────────────────────
## 
## Estimated Marginal Means of "Type":
## Treatment = nonchilled:
##  Type        Mean  S.E.   [95% CI of Mean]
##  Quebec      35.33 (1.75) [31.86, 38.81]  
##  Mississippi 25.95 (1.75) [22.48, 29.43]  
## 
## Treatment = chilled:
##  Type        Mean  S.E.   [95% CI of Mean]
##  Quebec      31.75 (1.75) [28.28, 35.23]  
##  Mississippi 15.81 (1.75) [12.34, 19.29]  
## 
## Estimated means use an equally weighted average. 
## 
## Pairwise Comparisons of "Type":
## Treatment = nonchilled:
##  Contrast             b      S.E.   df t     p         Cohen's d [95% CI]  
##  Mississippi - Quebec  -9.38 (2.47) 80 -3.80 <.001 *** -1.17 [-1.79, -0.56]
## 
## Treatment = chilled:
##  Contrast             b      S.E.   df t     p         Cohen's d [95% CI]  
##  Mississippi - Quebec -15.94 (2.47) 80 -6.45 <.001 *** -1.99 [-2.60, -1.38]
## 
## SD_pooled for computing Cohen’s d: 8.01 
## No need to adjust p values. 
## Disclaimer (about Cohen’s d):
## There is considerable disagreement on how to compute Cohen’s d.
## You should not take the above output as the only right results.
## You are completely responsible for setting the "sd.pooled".

效应值

lm(Sepal.Length~Sepal.Width+Species,data=iris) %>% 
  effectsize::effectsize()
## # Standardization method: refit
## 
## Parameter         | Coefficient (std.) |         95% CI
## -------------------------------------------------------
## (Intercept)       |              -1.37 | [-1.55, -1.20]
## Sepal.Width       |               0.42 | [ 0.31,  0.53]
## Speciesversicolor |               1.76 | [ 1.49,  2.03]
## Speciesvirginica  |               2.35 | [ 2.11,  2.59]
aov(Sepal.Length~Sepal.Width+Species,data=iris) %>% 
  effectsize::effectsize()
## Parameter   | Eta2 (partial) |       90% CI
## -------------------------------------------
## Sepal.Width |           0.05 | [0.01, 0.12]
## Species     |           0.72 | [0.66, 0.77]
cor.test(iris$Sepal.Length,iris$Sepal.Width) %>% 
  effectsize::effectsize()
## r     |        95% CI
## ---------------------
## -0.12 | [-0.27, 0.04]
BayesFactor::correlationBF(iris$Sepal.Length,iris$Sepal.Width) %>% 
  effectsize(test=NULL)
## Warning in .recacheSubclasses(def@className, def, env): undefined subclass
## "numericVector" of class "Mnumeric"; definition not updated
## Summary of Posterior Distribution
## 
## Parameter | Median |        95% CI
## ----------------------------------
## rho       |  -0.11 | [-0.27, 0.04]
aov(mpg~wt*disp,data=mtcars) %>% effectsize::effectsize()
## Parameter | Eta2 (partial) |       90% CI
## -----------------------------------------
## wt        |           0.83 | [0.73, 0.89]
## disp      |           0.16 | [0.01, 0.36]
## wt:disp   |           0.32 | [0.10, 0.51]
aov(mpg~wt*disp,data=mtcars) %>% broom::tidy()
## # A tibble: 4 x 6
##   term         df sumsq meansq statistic   p.value
##   <chr>     <dbl> <dbl>  <dbl>     <dbl>     <dbl>
## 1 wt            1 848.  848.      141.    1.96e-12
## 2 disp          1  31.6  31.6       5.25  2.97e- 2
## 3 wt:disp       1  77.9  77.9      12.9   1.23e- 3
## 4 Residuals    28 169.    6.03     NA    NA
BayesFactor::anovaBF(Sepal.Length~Species,data=iris)
## Bayes factor analysis
## --------------
## [1] Species : 1.425293e+28 ±0%
## 
## Against denominator:
##   Intercept only 
## ---
## Bayes factor type: BFlinearModel, JZS
BayesFactor::lmBF(Sepal.Length~Petal.Width+Sepal.Width,data=iris)
## Bayes factor analysis
## --------------
## [1] Petal.Width + Sepal.Width : 3.376503e+36 ±0.01%
## 
## Against denominator:
##   Intercept only 
## ---
## Bayes factor type: BFlinearModel, JZS
BayesFactor::regressionBF(Sepal.Length~Petal.Width+Sepal.Width,data=iris)
## Bayes factor analysis
## --------------
## [1] Petal.Width               : 6.575863e+33 ±0.01%
## [2] Sepal.Width               : 0.4544437    ±0%
## [3] Petal.Width + Sepal.Width : 3.376503e+36 ±0.01%
## 
## Against denominator:
##   Intercept only 
## ---
## Bayes factor type: BFlinearModel, JZS
mtcars %>% tidyBF::bf_oneway_anova(x=cyl,y=mpg) %>%
  select(log_e_bf10,r2)
## # A tibble: 6 x 2
##   log_e_bf10    r2
##        <dbl> <dbl>
## 1       14.9 0.712
## 2       14.9 0.712
## 3       14.9 0.712
## 4       14.9 0.712
## 5       14.9 0.712
## 6       14.9 0.712

颜色

show_colors(colors=viridis::viridis_pal()(10))

show_colors(colors=ggsci::pal_aaas()(10))

show_colors(RColorBrewer::brewer.pal(name="Set2", n=8))

show_colors(RGB(0, 0, 255))

show_colors("#0000FF")

osummer3148
关注
专栏目录
bruceR:这是CRAN R软件包系统信息库的只读镜像。 bruceR —广泛使用的便捷高效的R函数。 主页
04-11
布鲁斯 BR oadlyüsefulÇonvenient和E fficient R里面的函数即BR荷兰国际集团ÜSERSÇoncise和E legant R数据分析。 该软件包包括易于使用的功能，用于（1）基本的R编程（例如，将工作目录设置为当前文件所在的位置，以丰富的格式和颜色打印字符串）； （2）多元计算（例如，使用反向计分来计算规模总和/均值/ ...）； （3）可靠性和因素分析； （4）描述性统计和相关分析； （5）方差多因素分析（ANOVA），简单效应分析和事后多重比较； （6）回归模型的整洁报告（控制台和Word）； （七）调解与适度分析； （8）用于统计和图形的附加工具箱。 作者 电子邮件： 网站： | | 引文 Bao，H.-W.-S. （2021年）。 bruceR：广泛有用的便捷高效的R函数。 R软件包版本0.xx https://CRAN.R-proj
R语言 bruceR包安装及常见错误解决
半糖奶茶加珍珠
04-16 2769
推荐大家看如下两篇文章，里面包含了如何安装bruceR这个包，以及安装时常见的错误和解决方法的讲解 参考文档：如何安装（点它！） 参考文档：安装常见问题的解决方法（安装有问题就点它！） 大神已经写的很清楚了，接下来我来说下我碰到的问题吧 起初，我已经成功安装好了“devtools”、“tidyverse”、“ggstatsplot”这三个包，但是我没有进行update.packages包的更新，而...
R语言报错 | object ‘integral‘ not found whilst loading name
最新发布
Rita_rr的博客
08-29 673
加载R包时出现报错Registered S3 method overwritten by 'htmlwidgets': method from print.htmlwidget tools:rstudio Error: package or namespace load failed for ‘Seurat’: object 'integral' not found whilst loading namespace 'spatstat.core'
bruceR 开源项目安装与使用指南
gitblog_00837的博客
08-16 761
bruceR 开源项目安装与使用指南 bruceR???? BRoadly Useful Convenient and Efficient R functions that BRing Users Concise and Elegant R data analyses.项目地址:https://gitcode.com/gh_mirrors/br/bruceR 1. 项目目录结构及介绍 bruceR 是...
基于burceR包和mma包的多重中介效应分析学习笔记
2301_79584199的博客
05-25 1778
（本文章部分学习自https://www.zhihu.com/tardis/zm/art/376007591?
探索“bruceR”：您的高效数据分析助手
gitblog_00337的博客
08-16 716
探索“bruceR”：您的高效数据分析助手 bruceR???? BRoadly Useful Convenient and Efficient R functions that BRing Users Concise and Elegant R data analyses.项目地址:https://gitcode.com/gh_mirrors/br/bruceR 在繁忙的数据科学领域中寻找一款既强大...
brucer可以做cox的中介吗
10-27
Brucer是可以成为Cox的中介的。作为中介，Brucer将充当由Cox雇佣的公司与其他公司或个人之间的桥梁。Brucer将负责促进Cox与其他合作伙伴之间的沟通和交流，以促进业务合作和合同达成。作为一名中介，Brucer将拥有...
r语言bruceR结果怎么看
08-09
BruceR是R语言中一个用于绘制流程图的包，它基于igraph库实现了图形的绘制。当你运行一个BruceR函数后，通常会得到一个包含节点和边的流程图结构。查看结果一般分为以下几个步骤： 1. **查看图表数据**：`plot()` ...
BRUCE
10-24
BRUCE
R语言| 中介效应分析，Mediation包和BruceR包，循环Process函数
热门推荐
mocchan925的博客
04-24 3万+
目录Brief summary中介效应的检验分析方法Mediation包 Brief summary 网上关于中介效应的资料挺多的，这里就不对原理进行过多解释。 简单来说，我们分析自变量 X 对因变量 Y 产生的影响，如果变量 X 通过影响变量 M 来影响变量 Y ，那么这个变量 M 就是中介变量。 例如租客 (X) 通过中介公司 (M) 找到合适的房子 (Y)，中介公司就在其中扮演了中介变量的角色，中介变量发挥的作用就称为中介效应了。 对于该分析我个人有一个认知，就是X和Y必须有关联。这个关联可以是数
Bruce
10-25
Bruce
探索数据之美：全面领略bruceR的强大魅力
gitblog_00062的博客
06-10 430
探索数据之美：全面领略bruceR的强大魅力 bruceR???? BRoadly Useful Convenient and Efficient R functions that BRing Users Concise and Elegant R data analyses.项目地址:https://gitcode.com/gh_mirrors/br/bruceR 在数据分析的广阔天地里，有一个工具...
R语言 生信 TCGA准备，TCGAbiolinks包安装存在的问题
aydayUP111的博客
09-22 3719
这是或许需要重启Rstudio就可以成功安装之前删除的dbplyr以及data.table这两个包了。这里按照R提示的把相应的包删除，我要删除的是dbplyr,data.table这两个包。三个R包：TCGAbiolinks，RTCGA，gdcRNAtools。若显示版本，则下载成功，之后可能出现再Rstudio里不显示的问题。删除后再加载TCGAbiolinks就会再R包界面出现该包了。不过我目前加载TCGAbiolinks时依然会有提示说。方便平台：Xena，第三方处理的，可能存在更新不及时。
caret教程05：类不平衡数据的处理
医学和生信笔记的博客
09-16 543
分类问题中结果变量的类不平衡很常见，一个类明显多于另一种，这样的结果会产生很大影响。SMOTEROSEDMwRROSE除此之外，有些算法在自带权重选项，可以给类少的类别配上更高的权重，这样也可以减少类不平衡带来的影响。关于这几种方法的详细解释，大家可以自行了解，我们主要演示caret中的实现方法。
bruceR包Freq()函数频率百分比统计
Mrrunsen的博客
09-13 217
我想统计一下不同学历水平被试的样本数和占比。用Freq()函数。我想考察违约与未违约人群的样本数和比例。
OneR算法简单应用
04-12
python 数据挖掘入门，熟悉OneR算法实现分类，了解OneR算法的基本操作
java 反编译 报错_java反编译后再编译成class时提示缺少包
weixin_36308404的博客
02-16 1461
问题描述我下载了一个对方的Jar程序,可以正常运行,但是我想修改程序里面的一些内容,比如其程序中有一个函数public void setNumTarget(final long numTarget) {this.numTarget = numTarget;}我想让其变成类似这样的(当然这只是其中一处想要修改的地方,还有几处稍微复杂点的)public void setNumTarget(final ...
医学案例|单因素重复测量方差
m0_37228052的博客
07-28 283
为研究某诱导方法对患者收缩压的影响，将手术要求基本相同的15名患者，在手术过程中采用A麻醉诱导方法，并在T0（诱导前）、T1、T2、T3、T4五个时刻测量患者的收缩压，试着进行方差分析。
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值