Comments for Analytics Made Skeezy

Comment on What is this site? by tyrone

tyrone — Fri, 17 May 2013 18:14:00 +0000

all i want for my birthday is a big black juicy footlong popsicle

Comment on What is this site? by tyrone

tyrone — Fri, 17 May 2013 18:12:00 +0000

this is tyrone and i don’t give a crap

Comment on Forecasting Continued: Using Simulation to Create Prediction Intervals Around Holt-Winters by John Foreman

John Foreman — Fri, 26 Apr 2013 00:23:00 +0000

Thanks for catching the column B/C typo! Fixed it now I think.

And you’re totally right that you can just loop the macro in VBA. I just didn’t want to force people to write code who were gun-shy.

Comment on Forecasting Continued: Using Simulation to Create Prediction Intervals Around Holt-Winters by jaedee

jaedee — Thu, 25 Apr 2013 22:43:00 +0000

Great post, glad to see a new one.

A small improvement and a (possible) correction:

1. Instead of just running your simulations by holding the option+command for minutes (it does take a long time), you can record the macros, then add a loop, i.e.:

Sub Autorun_My_Freaking_Macros()
‘
‘
‘ Keyboard Shortcut: Option+Cmd+z
‘
Dim i, j, k As Long

i = 1
j = 22
For i = 1 To Range(“k2″)

Calculate

Next i
End Sub

Then just set k2 to the number of simulations (i.e. 1000). You can also just hard code the numer by replacing the Range(“k2″) with a number. You can reference your macros in the new one, or add the relevant code before and after.

2. Correction (?): I believe you want to copy the demand in column C, not B. (See the step 2 where you “Copied the 2013 simulated demand data in column B”).

Comment on Check Yo Self: 5 Things You Should Know About Data Science (Author Note) by john4man

john4man — Wed, 17 Apr 2013 12:58:00 +0000

I agree that the term is rather silly. In general, I prefer the term analytics to data science. I imagine it comes from the idea that many data scientists have skills in teasing out insight from very large transactional datasets.

Comment on Check Yo Self: 5 Things You Should Know About Data Science (Author Note) by Stephen Hardy

Stephen Hardy — Wed, 17 Apr 2013 10:43:00 +0000

Data scientist seems like a bit of a tautology. What does a scientist test their hypothesis with if it isn’t data?

Comment on Optimal Blending of Cocaine by Alastair McDermott

Alastair McDermott — Tue, 16 Apr 2013 22:38:00 +0000

There’s very much a book in this current narrative, love it. It’s a great intro for a non data scientist like me (I’m learning more about what’s possible, rather than implementation details).

Comment on Optimal Blending of Cocaine by john4man

john4man — Tue, 16 Apr 2013 22:32:00 +0000

In the book I’m writing for Wiley (coming out the this Fall) I work through a bunch of orange juice blending problems (http://www.businessweek.com/articles/2013-01-31/coke-engineers-its-orange-juice-with-an-algorithm) and do a more thorough job of explaining simplex etc.

So that’s one analogy that I’m actually going to use!

Comment on Introduction to Machine Learning: Logistic Regression for Predicting Bad Trips by john4man

john4man — Tue, 16 Apr 2013 22:29:00 +0000

Yes, it’s standard practice to consider degrees of freedom and drop a column when doing dummy variables since it’s not providing any additional information.

Comment on Introduction to Machine Learning: Logistic Regression for Predicting Bad Trips by Alastair McDermott

Alastair McDermott — Tue, 16 Apr 2013 21:14:00 +0000

“Except that since you’ve got 7 drug options, we only need columns for 6
of them. If they’re all 0, then we’ll know that the 7th, absent column
is a 1.”

Is this regular practice in datamining? From a software perspective including data by implication like this would be considered very bad practice.