R: Experiment 1: With varying numbers of samples

Descriptive Statistics	First 100	First 1K	First 10K	First 100K
Mean	0.02000071	0.020000066	0.020000004	0.02
Standard Error	2.12714E-06	7.53406E-07	2.51164E-07	9.69855E-08
Median	0.020005	0.020004	0.020004	0.020004
Mode	0.020005	0.020005	0.020005	0.020005
Standard Deviation	2.12714E-05	2.38248E-05	2.51164E-05	3.06695E-05
Sample Variance	4.52471E-10	5.67621E-10	6.30831E-10	9.40618E-10
Kurtosis	28.87137928	21.46428225	19.07376827	12.23083198
Skewness	-5.453831468	-4.509853108	-3.831289593	-2.003065575
Range	0.000135	0.000252	0.000277	0.000374
Minimum	0.01988	0.019872	0.019868	0.019815
Maximum	0.020015	0.020124	0.020145	0.020189
Sum	2.000071	20.000066	200.000044	1999.999951
Count	100	1000	10000	100000
Confidence Level(95.0%)	4.2207E-06	1.47844E-06	4.92331E-07	1.9009E-07

foo<-function(n){
v <-1:12
v[1]=mean(To_Chip_RTP_delay[1:n])
v[2]=std.error(To_Chip_RTP_delay[1:n])
v[3]=names(sort(-table(To_Chip_RTP_delay[1:n])))[1]
v[4]=sd(To_Chip_RTP_delay[1:n])
v[5]=var(To_Chip_RTP_delay[1:n])
v[6]=kurtosis(To_Chip_RTP_delay[1:n])
v[7]=skewness(To_Chip_RTP_delay[1:n])
v[8]=min(To_Chip_RTP_delay[1:n])
v[9]=max(To_Chip_RTP_delay[1:n])
v[10]=sum(To_Chip_RTP_delay[1:n])
v[11]=length(To_Chip_RTP_delay[1:n])
v[12]=qnorm(0.965)*std.error(To_Chip_RTP_delay[1:n])
return(v)}
seq1<-c(foo(100),foo(1000),foo(10000),foo(100000))
mat1<-matrix(seq1,  ncol=4)

fee<-function(n) {foo(To_Chip_RTP_delay, 10^n)}

lapply(c(2:5), fee)
[[1]] [1] "0.0200006800000119"   "2.12697347407497e-06" "0.0200049999984913“
       [4] "2.12697347407497e-05" "4.52401615941855e-10" "30.3672958382318“ 

---

Transcript

Now, it's really easy to compute the statistics over the first hundred, thousand, ten thousand, hundred thousand.

Well, yes just define yourself a function to compute the statistics we want, and then 

calling the function to be applied to the function with a hundred, a thousand, ten thousand, or a hundred thousand samples will produce the columns. We are set.  Off we go!