Appendix

library(robustHD)
library(readxl)

P/Net Sales

Beroende variabel: P/Net Sales
Oberoende Variabel: Ledare(1)

Deskriptiv statistik för P/Net Sales

Psales <- read_excel("~/Documents/analysis/psales.xlsx")
summary(Psales$`P/Net sales`)

    Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
 0.00016  0.59745  1.31189  2.86302  2.60889 42.82621 

Parametriskt test: T-test av P/Net Sales

#H0: P/Net Sales ar lika for ledare och foljare
#t-test
t.test(Psales$`P/Net sales` ~ Psales$Leader, mu=0, alt="two.sided")

    Welch Two Sample t-test

data:  Psales$`P/Net sales` by Psales$Leader
t = 2.5523, df = 73.889, p-value = 0.01277
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 0.3113215 2.5278742
sample estimates:
mean in group 0 mean in group 1 
       2.939759        1.520162 

boxplot(Psales$`P/Net sales` ~ Psales$Leader)

Parametriskt test: Winsoriserat(5%) t-test av ‘P/Net Sales’

t.test(winsorize(Psales$`P/Net sales`) ~Psales$Leader, mu=0, alt="two.sided")

    Welch Two Sample t-test

data:  winsorize(Psales$`P/Net sales`) by Psales$Leader
t = 0.48236, df = 9.3824, p-value = 0.6406
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.5379892  0.8319103
sample estimates:
mean in group 0 mean in group 1 
       1.667122        1.520162 

boxplot(winsorize(Psales$`P/Net sales`)~Psales$Leader)

Icke-parametriskt test: Mann-Whitney U-test

wilcox.test(Psales$`P/Net sales`~Psales$Leader, mu=0, alt="less", conf.int=T, conf.level=0.95, paired=F, exact=T,correct=T)

    Wilcoxon rank sum test

data:  Psales$`P/Net sales` by Psales$Leader
W = 530, p-value = 0.403
alternative hypothesis: true location shift is less than 0
95 percent confidence interval:
      -Inf 0.7360202
sample estimates:
difference in location 
           -0.08125674 

P/EBITDA

Beroende variabel: P/EBITDA
Oberoende Variabel: Ledare(1)

Deskriptiv statistik för P/EBITDA

Pebitda <- read_excel("~/Documents/analysis/pebitda.xlsx")
summary(Pebitda$`P/EBITDA`)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
-2.4066  0.4552  0.8967  2.5901  1.6212 64.7410 

Parametriskt test: T-test av P/EBITDA

#t-test
t.test(Pebitda$`P/EBITDA` ~Pebitda$Leader, mu=0, alt="two.sided")

    Welch Two Sample t-test

data:  Pebitda$`P/EBITDA` by Pebitda$Leader
t = 1.4078, df = 87.192, p-value = 0.1628
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.4305258  2.5213128
sample estimates:
mean in group 0 mean in group 1 
       2.641281        1.595888 

boxplot(Pebitda$`P/EBITDA` ~Pebitda$Leader)

Parametriskt test: Winsoriserat(5%) t-test av ‘P/Net Sales’

t.test(winsorize(Pebitda$`P/EBITDA`) ~Pebitda$Leader, mu=0, alt="two.sided")

    Welch Two Sample t-test

data:  winsorize(Pebitda$`P/EBITDA`) by Pebitda$Leader
t = -1.4831, df = 6.8288, p-value = 0.1827
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -1.0866944  0.2515793
sample estimates:
mean in group 0 mean in group 1 
       1.081034        1.498591 

boxplot(winsorize(Pebitda$`P/EBITDA`) ~Pebitda$Leader)

Icke-parametriskt test: Mann-Whitney U-test

wilcox.test(Pebitda$`P/EBITDA`~Pebitda$Leader, mu=0, alt="greater", conf.int=T, conf.level=0.95, paired=F, exact=T,correct=T)

    Wilcoxon rank sum test

data:  Pebitda$`P/EBITDA` by Pebitda$Leader
W = 325, p-value = 0.9202
alternative hypothesis: true location shift is greater than 0
95 percent confidence interval:
 -0.9488513        Inf
sample estimates:
difference in location 
            -0.4392584 

P/EPS

Beroende variabel: P/EPS
Oberoende Variabel: Ledare(1)

Deskriptiv statistik:

Peps <- read_excel("~/Documents/analysis/peps.xlsx")
summary(Peps$`P/EPS`)

     Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
-138.1499    0.1547    0.7413    0.3884    1.5769   63.6761 

Parametriskt test: T-test av P/EPS

t.test(Peps$`P/EPS` ~Peps$Leader, mu=0, alt="two.sided")

    Welch Two Sample t-test

data:  Peps$`P/EPS` by Peps$Leader
t = -0.46903, df = 141.25, p-value = 0.6398
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -3.038867  1.873408
sample estimates:
mean in group 0 mean in group 1 
      0.3541537       0.9368830 

boxplot(Peps$`P/EPS` ~Peps$Leader)

Parametriskt test: Winsoriserat(5%) t-test av ‘P/EPS’

t.test(winsorize(Peps$`P/EPS`) ~Peps$Leader, mu=0, alt="two.sided")

    Welch Two Sample t-test

data:  winsorize(Peps$`P/EPS`) by Peps$Leader
t = -0.027212, df = 9.3798, p-value = 0.9789
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.8329783  0.8130559
sample estimates:
mean in group 0 mean in group 1 
      0.8421871       0.8521483 

boxplot(winsorize(Peps$`P/EPS`) ~Peps$Leader)

Icke-parametriskt test: Mann-Whitney U-test

wilcox.test(Peps$`P/EPS`~Peps$Leader, mu=0, alt="two.sided", conf.int=T, conf.level=0.95, paired=F, exact=T,correct=T)

    Wilcoxon rank sum test

data:  Peps$`P/EPS` by Peps$Leader
W = 684, p-value = 0.7866
alternative hypothesis: true location shift is not equal to 0
95 percent confidence interval:
 -0.7913083  0.8234610
sample estimates:
difference in location 
             0.1206787