[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Manipulation of the distribution

From   "Martin Weiss" <>
To   <>
Subject   st: RE: Manipulation of the distribution
Date   Mon, 10 Nov 2008 13:29:48 +0100

Hard to figure out your email in my Outlook (weird formatting, of the code
in particular). You want to look at 

-ssc d rnd-
-h sample-

the former because you run Stata9: Stata 10.1 now comes with built-in random
number functionality. 

BTW, -ret list- is very much unnecessary in a loop. This command just serves
to acquaint you with the return list, but is of no use in the loop...


-----Original Message-----
[] On Behalf Of
Sent: Monday, November 10, 2008 1:11 PM
Subject: st: Manipulation of the distribution

HiI have a dataset of hourly prices. Prices are identified by hour, weekday,
week number, month and year. So within each day I have 24 hourly prices.
What I'd like to do is to manipulate the prices by changing the distribution
(changing the volatility of the prices following some assumptions on how the
market will be changed). Let's say I want to increase the daily price
volatility by for example increasing the standard deviation within each day.
We've figured out a way to do this systematically using a group variable and
directly changing the original standard deviation for each day. egen
group_var=group(day week)capture drop new_no1gen new_no1=no1sum group_var
forvalues i=1(1) `r(max)' {  sum no1 if group_var==`i'  ret list  local
org_mean `r(mean)'  local org_sd `r(sd)'  capture drop std_no1  gen
std_no1=(no1-`org_mean')/`org_sd' if group_var==`i'  sum std_no1 if
group_var==`i'  local new_mean `org_mean'  local new_sd 1.2*`org_sd'
replace new_no1=`new_sd'*std_no1 + `new!
 _mean' if group_var==`i'  sum new_no1 if group_var==`i'  }However, using
this method will only systematically "stretch" the distribution so that the
originally highest prices increases and the lowest prices decreases. Making
the original distribution of prices more extreme.My trouble is to find a way
to change the prices by randomly drawing a "price change" for each hourly
price. I think what I need is to generate a variable for example one that is
normally distributed with a given mean and standard deviation. Then I want
to draw randomly from this distribution for each price in my dataset and add
this element to the original price in order to have a "new price".Well, I
have some trouble understanding how to actually generate new variables with
a given distribution, say a normally or a binomial distributed variable
(could really use som instruction on this). Second, I need to figure out how
to tell Stata to draw randomly from this new variable distribution (one for
each pri!
 ce observation) and then assigned the element drawn to the new price. 

(well this last thing would be easy as soon as I have the draws). I've
looked into for example the rnd command written by Joe Hilbe, and also some
other commands like normal, drawnormal ect (I'm running Stata 9), but I'm
far from "fluent in Stata", so I could need some help.ThanksLinn
Hold kontakten med Windows Live Messenger.
*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index