Math 300 , Spring 2008, Day 31, M, April 14Hit reload.After class...
Mini-exam due Wed.

Today, Cornell 4:30--5:30, in Bache Auditorium, Malott Hall
Prof. David Aldous of Berkeley
*“The Ten Top Things that Math Probability Tells You About the Real World.”*
Supposed to be accessible to undergraduates, and poking around for info about him, he seems to be a lively and engaged sort. His research is high powered, but he clearly likes very much connecting the math with the "real world."

Prof. Aldous' picture: Scroll down to 2nd from bottom: http://adwhiteprofessors.cornell.edu/visits2007.html
I'm taking the 3:00 van down to Cornell; Malott Hall (the math dept) is only about 2 blocks from the Sage Hall/Statler stop. Please consider joining me!    Cornell campus map  for the Statler--Malott area.

HW Try to Finish all the HW given on Friday. Day 30
Bring questions!
Next, Ash Ch. 5 (first 5.1, 5.2)

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =  = =
Functions of Random Variables, cont.:  See Day 29.  New: Density method, Day 29

Figuring out what x's map to a y, or what ranges of x's map to a range of y.  or what dx's map to a dy.

A graph of y =g(x) may help, going over and down from a y to see what x('s) map to it. (Ash's favorite method)  Think of the probability density for the x's piled along the x axis, the probability for the y's turned sideways and piled along the y-axis.
Examples: Uniform to e, and  Lognormal, in Excel:  fX(x)dx =area in dx interval = area in dw interval = fW(w)dw.  (Dot to dot is an interval)


Handouts from last time:

--Two sided handout, the Lognormal distribution (W is lognormal, X = ln(W) is normal.  W = eX)
    Skewed data often "turns normal" when you do a logarithm to it.  If so, the original data is "lognormal."  
  This handout  has homework problems A-D in the corner.

Other details:
-- One sided handout, Changing variables when there's a Y = -X type situation.   That is CDF method. 
            Density, just drop the minus in the derivative (Day 29, additional technique c).  Try it with the handout W = -X normal.

When the transformation function is not 1-1,  two (or more) x's map onto the same y.  (Day 29, additional technique d)
      The Lognormal handout, backside, bottom, models the Density method. 
   The CDF method may give a double inequality, to evaluate using the X distribution..
            Cf. 6-sided die, W = (X-3)2.  FW(w) = P(3 - sqrt(w) < x < 3+ sqrt(w))  Day29

Applications (lognormal or other transformations):  Xeroxed pages outside my door, "Re-expressing...."
      Just reading the  NY Times this weekend I noticed two highly skewed distributions;  half of the $$ given to a charity tend to come from 10% of the donors.  Half of the revenues to gambling casinos come from 5% of the gamblers.  The numbers are wrong but the  ballpark idea is there.  The "80-20" rule, which I learned this way:  The first 80% of a job can be done in 20% of the time; the remaining 20% will take the other 80% of the time to acheive completeness or perfection.  How (not) to use it.  If something is worth doing, it's worth doing badly...



Sievers home  Math300-Sp08/Day8p31.htm  2pm 4/14/08
This page belongs to Sally Sievers who is solely responsible for its content. Please see our statement of responsibility.