[Stef Appel]

Stef:
simply replace your values in A (or import your 584-number variable and name it -A- in Stata.dta) and run the rest of the code.

What I actually did, the result is the same. Is something wrong?

[Carlo Lazzaro]

Stef:
you repeated the old code and, being -seed- set01 you get te very same results.
Now you have your origina data for -A-, instead.
Hence the previous code should be tweaked a bit, just like in the following toy-example.

import excel "C:\Users\user\Desktop\StefSample.xlsx", sheet("Foglio1") clear

set seed 1

 g double mean=.
(584 missing values generated)


quietly forvalues i = 1/100 {
preserve
sample 30 if A !=., count
summ A
scalar Am = r(mean)
restore
replace mean = Am in `i'
 }

sum A mean

[Stef Appel]

Stef:
you repeated the old code and, being -seed- set01 you get te very same results.
Now you have your origina data for -A-, instead.
Hence the previous code should be tweaked a bit, just like in the following toy-example.

import excel "C:\Users\user\Desktop\StefSample.xlsx", sheet("Foglio1") clear

set seed 1

g double mean=.
(584 missing values generated)


quietly forvalues i = 1/100 {
preserve
sample 30 if A !=., count
summ A
scalar Am = r(mean)
restore
replace mean = Am in `i'
}

sum A mean

YES!!! Carlo you are a great master. I wish you all the best!

Best regards,
Stefan.

[Stef Appel]

Stef:
you repeated the old code and, being -seed- set01 you get te very same results.
Now you have your origina data for -A-, instead.
Hence the previous code should be tweaked a bit, just like in the following toy-example.

import excel "C:\Users\user\Desktop\StefSample.xlsx", sheet("Foglio1") clear

set seed 1

g double mean=.
(584 missing values generated)


quietly forvalues i = 1/100 {
preserve
sample 30 if A !=., count
summ A
scalar Am = r(mean)
restore
replace mean = Am in `i'
}

sum A mean

Carlo, I have one more question.
Is there random sample by picking up 30 numbers?
What I see, that computer gives me all the time the same result. but, in reality it should fluctuate a little bit, because by random choice numbers should vary.
Do you know what I mean?
And if I try for instance the next code, all the time the programm gives me the same result.

set seed 1
g double mean=.
quietly forvalues i = 1/1 {
preserve sample 1 if A !=., count
summ A
scalar Am = r(mean)
restore replace mean = Am in `i'
}
su A mean

[Carlo Lazzaro]

Stef:
if you remove -set seed 1- from your code, you'll get different results every time you run it.

[Stef Appel]

Stef:
if you remove -set seed 1- from your code, you'll get different results every time you run it.

Thanks a lot, Carlo.

Best regards,
Stef.

[Stef Appel]

Stef:
if you remove -set seed 1- from your code, you'll get different results every time you run it.

Carlo:
when the programme picks up the numbers, 30 from 584. Is there 30 different numbers? or could it be that the programme picks up twice or three times the same number?
For instance, I need to do it also for 500 numbers. All 500 numbers will be different, without duplications?

Best regards,
Stef.

[Carlo Lazzaro]

Stef:
yes, they're different with -sample-;
no, n_plications may occurr with -bsample- (but this is not the command you selected).

[Stef Appel]

Stef:
yes, they're different with -sample-;
no, n_plications may occurr with -bsample- (but this is not the command you selected).

Carlo:
Is it possible with Stata to compute correlation?
I have 584 numbers and need to get correlations between all stocks, and then to compute the mean of them? Is it possible?



Best regards.
Stef.

[Carlo Lazzaro]

Stef:
yes.
See, among others, -help correlate-.

[Marcos Almeida]

Just as a side note, if you need to perform this operation 100 times, you probably need to make sure every one may reproduce it , hence you should think about setting seeds.