What kind of transformation procedure need to applied for this kind of distribution?

[Reeza]

It's hard to see what your doing now, since the forum doesn't sort the responses. So no way to tell which reply goes with which post. In the future please just do one response. This isn't your fault - limitation of the forum. 

Did you post your original distribution? Also, what is the variable, in laymans terms. Context can help with deciding what type of transformation to use, and there are certain standard transformation in diff industries. 

[Ksharp]

Or you could check Box-Cox transformation . 
PROC TRANSREG  or PROC MCMC can do that.
Check example of them in documentation.

[mantubiradar19]

Let me check these two! Xie Xie

[mantubiradar19]

I found this code from SAS documentation. I have 10,000 observations. So I need to use i=1 to 10,000 right?

data x;
call streaminit(17);
z = 0;
do i = 1 to 10000;
FASTING_GLUCOSE = rand('lognormal');
output;
end;
run;

proc transreg maxiter=0 nozeroconstant;
model BoxCox(FASTING_GLUCOSE) = identity(z);
output;
run;

proc univariate noprint;
histogram FASTING_GLUCOSE tFASTING_GLUCOSE;
run;

[Ksharp]

Interesting. I got this:



data x;
call streaminit(17);
do i = 1 to 10000;
FASTING_GLUCOSE = rand('lognormal');
x=rand('normal');
output;
end;
run;
proc transreg data=x details pbo noprint;
model boxcox(FASTING_GLUCOSE  / convenient lambda=-2 to 2 by 0.01) = identity(x);
output out=trans;
run;

proc univariate data=trans plots;
histogram FASTING_GLUCOSE tFASTING_GLUCOSE ;
run;

[mantubiradar19]

Thank you very much! Let me try this and get back to you! Cheers

[Ksharp]

I also found if you LOG10(x), you can also get what you want.

[mantubiradar19]

I tried the log10 transformation but still the distribution is not so
normal!

##- Please type your reply above this line. Simple formatting, no
attachments. -##

[mantubiradar19]

In layman terms I'm dealing with a contineous variable! Here is the distrubution before and after the log transformation!

[Reeza]

---

@mantubiradar19 wrote:

In layman terms I'm dealing with a contineous variable! Here is the distrubution before and after the log transformation!

---

What type of continuous variable, blood pressure, stock price, number of asteroids heading towards earth? 

[mantubiradar19]

These are the Fasting glucose values!